CA2694451A1 - Cluster and discriminant analysis for vehicles detection - Google Patents
Cluster and discriminant analysis for vehicles detection Download PDFInfo
- Publication number
- CA2694451A1 CA2694451A1 CA2694451A CA2694451A CA2694451A1 CA 2694451 A1 CA2694451 A1 CA 2694451A1 CA 2694451 A CA2694451 A CA 2694451A CA 2694451 A CA2694451 A CA 2694451A CA 2694451 A1 CA2694451 A1 CA 2694451A1
- Authority
- CA
- Canada
- Prior art keywords
- cluster
- data
- cndot
- new
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/36—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source using liquid crystals
- G09G3/3607—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source using liquid crystals for displaying colours or for displaying grey scales with a specific pixel layout, e.g. using sub-pixels
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/3406—Control of illumination source
- G09G3/342—Control of illumination source using several illumination sources separately controlled corresponding to different display panel areas, e.g. along one dimension such as lines
- G09G3/3426—Control of illumination source using several illumination sources separately controlled corresponding to different display panel areas, e.g. along one dimension such as lines the different display panel areas being distributed in two dimensions, e.g. matrix
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/22—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters using controlled light sources
- G09G3/30—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters using controlled light sources using electroluminescent panels
- G09G3/32—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters using controlled light sources using electroluminescent panels semiconductive, e.g. using light-emitting diodes [LED]
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/3406—Control of illumination source
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/3406—Control of illumination source
- G09G3/3413—Details of control of colour illumination sources
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/3433—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source using light modulating elements actuated by an electric field and being other than liquid crystal devices and electrochromic devices
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/36—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source using liquid crystals
- G09G3/3611—Control of matrices with row and column drivers
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2320/00—Control of display operating conditions
- G09G2320/06—Adjustment of display parameters
- G09G2320/0626—Adjustment of display parameters for control of overall brightness
- G09G2320/0646—Modulation of illumination source brightness and image signal correlated to each other
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2320/00—Control of display operating conditions
- G09G2320/06—Adjustment of display parameters
- G09G2320/0666—Adjustment of display parameters for control of colour parameters, e.g. colour temperature
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2360/00—Aspects of the architecture of display systems
- G09G2360/16—Calculation or use of calculated indices related to luminance levels in display data
Abstract
A method is provided herein for determining and recognizing types of vehicles passing a cheek point.
The method takes advantage of an EM algorithm which is up-loaded into a CPU
and which processes data of the vehicles which drive past a checkpoint, the data being representative of essential characteristics of vehicles to produce an output model of the traffic volumes of the various types of vehicles. This model enables the forecasting of future road maintenance costs and the, planning and designing of future road networks.
The method takes advantage of an EM algorithm which is up-loaded into a CPU
and which processes data of the vehicles which drive past a checkpoint, the data being representative of essential characteristics of vehicles to produce an output model of the traffic volumes of the various types of vehicles. This model enables the forecasting of future road maintenance costs and the, planning and designing of future road networks.
Description
C
Tide, CLUSTER AND DISCRIMINANT ANALYSIS FOR VEHICLES DETECTION
100011 This invention relates to a system and method for cluster analysis for vehicle iden0catiort and claims priority of application 61/154866, filed 02/24/2009, the entire content of which are incorporated herein by reference
Tide, CLUSTER AND DISCRIMINANT ANALYSIS FOR VEHICLES DETECTION
100011 This invention relates to a system and method for cluster analysis for vehicle iden0catiort and claims priority of application 61/154866, filed 02/24/2009, the entire content of which are incorporated herein by reference
(0002) It is very useful to build an automatic computer system to recognize the types of vehicles passing a checkpoint given some easy-to-get data about the vehicles, such as the distances between axles, the weights on each axle. Such a system has many applications, for example, in monitoring traffic volumes and identifies the type of vehicle, which will be helpful in budgeting road maintenance costs.
[0003] The simplest clustering technique is the K-means clustering- However, K-means clustering requires that the users supply with a number of clusters. X-means clustering may be an alternative method since it can detect the number of clusters with some simple criteria, but X-means would introduce more severe local mode problem.
BACKGROUND INFORMATION
BACKGROUND INFORMATION
[0004] The partitioning of large data sets into similar subsets (Cluster Analysis) is an important statistical technique used in many fields (data 'mining, machine learning, bioinfonrnatics, and pattern. recognition and image analysis). In traffic research, it is useful both to determine and to recognize the types of vehicles passing a checkpoint.
Traffic data collection systems would collect data (e. g.., vehicle length, distances between axles, weights on axles) and such data may be used to determine and recognize vehicle types in high volume traffic, monitoring traffic volumes of various types of vehicles forecasting future road maintenance costs and planning and design of future road networks..
Traffic data collection systems would collect data (e. g.., vehicle length, distances between axles, weights on axles) and such data may be used to determine and recognize vehicle types in high volume traffic, monitoring traffic volumes of various types of vehicles forecasting future road maintenance costs and planning and design of future road networks..
[0005) ,The consequence of such determination and recognition of vehicle types in high volume traffic has many applications, C.&,, monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks.
DESCRIPTION OF THE INVENTION
AIMS OF THE INVENTION
DESCRIPTION OF THE INVENTION
AIMS OF THE INVENTION
(0006] A main aim of the present invention is to develop a better methodology for cluster analysis with application to the problem of vehicle detection and determination of its type as noted above.
[0007) Another aim of the present invention is to provide a new method to overcome potential. problems by merging similar clusters after running X-means clustering.
[0068] Another aim of the invention is to provide better methodology for cluster analysis with application to the vehicle detection problem.
STATEMENT OF INVENTION
(0009] One aspect of the present invention provides a method of determining and recognizing -the' types of vehicles passing a checkpoint by collecting vehicle data (e.g., vehicle length, distances between axles, weights on axles) and using that data to determine and recognize vehicle types, particularly in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks, the method comprising: uploading a computer program into a CPU, the computer program comprising an EM algorithm as particularly described in the specification herein, the algorithm including data representations, of essential characteristics vehicles as they drive past the checkpoint; entering such measured characteristics vehicles as they travel past the checkpoint into that CPU; and deriving an output from that CPU, and thereby determining and recognizing the types of vehicles passing the checkpoint, particularly in high volume traffic, for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks.
[00101 Another aspect of the present invention provides an apparatus comprising the combination of, a CPU: and a computer program which has been uploaded into said CPU, the computer program comprising an EM algorithm as particularly described in the specification herein, the algorithm including data representations of essential characteristics of vehicles.
[00'11) It has been found according to aspects of the present invention, that there are correlations between different variables. This invention proposes to avoid the problem which arises by using the Euclidean distance, since data may be assigned to the wrong centrolds. The present invention seeks to overcome this problem by replacing the Euclidean distance with the Mahalanobis distance-(00121 The following description provides examples of methods of aspects of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
In the accompanying drawings, [0013.) Fig 1 is .a graph showing some data points derived from traffic data collection systems which have collected data (e. g.., vehicle length, distances between axles, weights on axles, etc.), which data may be used to determine and recognize vehicle types in high volume traffic, but in which the cluster points are incorrectly clustered [0014] Fig 2 is a graph showing some data points derived from traffic data collection systems which have collected data (a. g.., vehicle length, distances between mules, weights on axles, etc.), which data may be used to determine and recognize Vehicle types in high volume traffic, but in which the cluster points are correctly clustered [0015] Fig 3 is a graph showing. some data points derived from traffic data collection systems which have collected data, (e. S.., vehicle length, distances between ,axles, weights on axles, etc.), which data may be used to deter-mine and recognize vehicle types in high volume traffic, where the X-means is used to cluster these points, using Euclidean distances, which is not correct.
[0016] one explanation of the results plotted in Figures I, 2 and 3 is because the X-means algorithin'does not permit returning back to re-cluster the data, since it runs a local K-means for each pair of "children". The K means is local in that the "children"
are fighting each other for the points in the "parent's"region and in no others. All the points from the other regions are ignored.
[0017] This problem of local mode can be overcome, according to broad aspects of the present invention, by merging two regions which are close to each other after the X-means algorithm is run. If the model after merging has a higher BIC score than the model before merging, these regions will be merged. Otherwise, the original model is kept.
ASSUMPTIONS
(00IS].In the method of determining and recognizing the types of vehicles passing a checkpoint by collecting vehicle data (e ,g., vehicle length, distances between axles, weights on axles) according to aspects of the present invention, by plotting graphs for all the variables (axle spacing, weights, front bumping spacing, and rear bumper spacing) in the data set, each variable forms a pattern which is similar to a "Student's" t. distribution. Therefore, it will be assumed that each variable in the data set comes from a "Student's" t-distribution Since all the variables for a given data point must' be considered, it will be assumed that each data point forms a multivariate "Student" distribution. In statistics, a "multivariate Student distribution" is a multivariate generalization of the Students t-distribution.
[40l9f The Present invention will be further described by reference to method steps to be carried out [ 00020] FINDING PARAMETER VALUES WITH EM ALGORITHM METHOD STEPS
[00021] The setup for the method steps is that given "N" data points in a V-dimensional space, it is desired to find a set of "K" "multivariate Student's t-distribution" that best represents the distribution of all the data points. Without being bound by theory, it is believed that the given data set is an N * V matrix, where N stands for the number of data points and V stands for the number of variables for each data point.
[OOO22] DEFT TION OF TERMS
N = number of data points.
V number of variables.
K number of dusters.
k = the mean for kill cluster, each a vector of length V.
E k:- the covariance matrices for kth duster, each of size V* V.
X11.= the nth data point, which is a vector with length V.
S
P(k ! xa:) - the probability that xn Comes from cluster k.
p(k) = : the probability that a data point chosen randomly comes from cluster k.
P(xn) = the probability of finding a data point at position xn = the value of log likelihood of the estimated parameter set.
(00023] For simplicity, it is assumed that k is a diagonal matrix, i.e., a matrix whose non diagonal entries are all 0, and where the diagonal entries are the variances for each variable.
[00024) Three statistical methods are used for the method steps, which are are carried out to find pararncter values with the EM algorithm, splitting clusters using Principle Component Analysis (PCA), and comparing models by Bayesian Information Criterion (BIC).
[00025] X is the key to tktis method. We it is not desired to be limited by any particular theory, it is believed that it is necessary to find the best values for the parameters by maximizing the value of I X. This method maximizes the posterior probability of the parameters if the specific priors are given.
[00026) The -steps are described as follows [00027] Step I
[00028] Set the=starting vahies forthe pks, Ek s, P (k). The method to obtain these values is by means of splitting clusters using PCA as follows:
[00029) The setup for this method is that, given some data points in one cluster (mother cluster),it is necessary to split these data points into two clusters (children clusters), using PCA.
PCA is mathestiatjcally deAned'as "an orthogonal linear transformation that transforms the data to anew coordinate system such that the greatest variance by any projection of the data comes to lie on the fast coordinate (called the first principal component), the second greatest 'variance on the second coordinate", and so on. The data set comprises an N * matrix. PCA
is .now performed an this data matrix. The standard deviations are now calculated of the principal components, namely, the square roots of the eigenvalues of the covariance matrix. The matrix is now calculated far variable loadings, namely, a matrix whose columns contain the eigenvectors.
In R, the is a built-in function called "preomp" which helps the calculations.
(00030] The terms used are defined as follows:
std = a vector contains the square roots of the eigenvalues of the covariance matrix.
Rotation = a matrix whose columns contain the eigenvectors Range: how far the data is to be split; the.value of range is usually between 1.5 and K = the mean for the mother cluster, each a vector of length V
E = the covariance matrix for the mother cluster, each of size 'P(M) = the probability that a data point chosen randomly comes from the ritother cluster (0003 i] Since the first principal component is the most important component, two vectors are created with length V from the first principal component. Two vectors are created since it is desired to to split the data into two clusters. The first element in. the first vector is the value of range (plus range) , and the other elements are all zero. The first element in the second vector is the value of -range (minus range) . and the other elements are all zero. The first vector is V 1, and the second vector is V 2. Consider V 1 , V2, and std to be matrices with one column. After the splitting is done, two meads are provided for two children clusters. The mean for the first children cluster is placid the mean for the second cluldren cluster }-2.The calculation for gland is as follows:
Pl li+ rotation W% (V 1 * Std) p2= 1,E+ rotation %* %o (V2 std) [000321 Here, k*k and "%.*%" different operations.
(00033]For example, a d a+d a d g j a*j +d*h+g*1 b*e -b*e but beh%*% k b*j +e*k+h*1 c f c*f c f i I c+j +f*k+l+l [0034] The covariance matrices for two children clusters would be the same as the mother cluster, and the probability that a data point chosen randomly comes from the children clusters would be half ofP(m).
[000351 In stmunary, for the first children cluster, mean = l , covariance matrix = E, and probability that a data point chosen randomly comes from this cluster = (I
/2)*P{m); for the second children cluster, mean 112, covariance matrix = E, and probability that a data point chosen randomly comes from this cluster = (1/2)*P(m).
[00030] There is one limitation about PCA. If PCA is performed on a given data matrix, PCA
requires 'the number of data points'to be larger than the number of variables.
If the number of data points is smaller than the number of variables, PCA will do nothing on the data matrix, and splitting will not happen [000361 Step 2-[0003.71 Given the values for the k's, F..k s, and P (k), and the data, of P
(xn 1 k , Ek) can be calculated. we assume all the variables in the data set form a multivariate Student's distribution, P (xn - k, Mk) is the multivariate Student's density, that is, rj td_'~
P (xul Pe. fir) {dr r (doh (detE) z I t (s) * (xa ^ Ne)F . E-A . (xn - pk}
where df w the degree of freedom, p T the number of variables, and detE = the determinant for [00037] One important thing about P (xn l k , Ek) is that the values of P (xn 1 Ak , FY often be very small as to underflow to zero. Thereforc,it is necessary to work with the logarithm ofP (xn 1 k Lk) instead of P(zn 1 k , Ek), that is 10$ P (X.1 uk= F-k) logr(df+p) - logr (d2) - log (rc)gz - log(df)Pz - log(det'E)z -iogl+\_) (xõ_1i*)Ts E r(,xx-Pk)J
[00038]After the value'of P (xn Ilk , Ek) is obtained, it becomes possible to calculate the value of P (xc) splitting P (xn) into its contribution from each of the K
multivariate Student's t-distributions, that is, P (xii) P (xn I Pk, Ek) P(k) k [00039] One problem may rise for P(xn), where it becomes necessary to calculate the sum of quantities. Some of these quantities may be so small that they underflow to zero. According to an aspect of the present invention, it has been found that one possible way to fix this problem is construct the quantities from their Logarithms. That is, store P (xn 1 k , F.k) P(i) in log P (xn l k , Zk), and let m, - max log (P (xn 10k , )-k)P(i)0..., log P(k))). Then the logarithm of the sum is computed as follows:
IagP (x,1) a log(m ,,) + log(l exp(log (P(; l1 L. Ef)F(t)) - mraa)) 1000401 Using the values of P ,(xn I k , Zk) and P (xn), the value of P(kl xn) and 3.:
P(klx,) = P(XRII~~=EK)P(k) P (xõ) A = tog (fl P(xõ )) _ X logP (x') 100041 ] Since the values of log P (x, ' I l k Ek) and log P (xn) can be computed in order to overcome the problem of underflow, it is possible to write P(k 1 xn) in terms of log P
(xn l k , Ek) and I. P (xn), p(klxx) = exp (IogP(xnl Lk.Er) + iogP(k) - logP(x,,)) [00042]By calculating P (ac I xn) for all values of k's and xõ's, it is now possible to obtain all.of the value's P (k I x,1}'s, and it now becomes possible to write P (k I xa)'s as a probability matrix of size N * K. Each row = one data point, and each column =
one cluster. Each element in the matrix = the value of P (k l xn), that is, the probability that a given data point comes from a specific cluster k. In the language of the EM
algorithm, this is called an expectation step or an E-step.
[00043] Step 3:
[00] Using P (k I Qs from step 2, the values of maximum likelihood estimates for k's is, Ek s and P(k)'s and for all values of k, can be calculated, that is , the values of k's , Zk's and Pik)'s that maximize the log likelihood function X. The maximum likelihood estimate for P(k) is easy to obtain-P(k) p (klx=,) 1000451 The process to calculate the maximum likelihood estimates for id's , rk's is very complex. For a given cluster k (k = t, 2, 3 ...K , ), the log likelihood function needed to maximize is as follows-A log (P (x l k=Ek)) * P(klx.) [00046j'11 is now necessary to find the values of pk and F..k'sfor k =1, 2, 3 ...K that maximize the above function.
[00047) Most of the programming languages have the build-in functions to calculate the values of the parameters that maximize a given function. For example, in R, we can use a build-in fimction called "DIM" can be used to calculate the maximum likelihood estimates for -k'S 'S. the language of the EM steps is called a maximization step or M-step.
[00045] Step 4;
[000491 Using the maxirnuri likelihood estimates for ttk's , F..k's and P(k}'s as the new pi's, F.k's and F(k}'s, repeat Step 2 and Step 3 until the value of X no longer changes.
(00050] After the clustering process, the final values for gl s , E, s and P(k)'s have been obtained for all values of k. A probability matrix whose entries, are the final values of F (k I Qs have also been obtained Given a data point,the corresponding row in the probability matrix for this data point can be found Then, it, is possible to determine which cluster most likely comes from by looking at the values of P
(k I ys from this row. The column index which produces the largest value of P (k I xn) is the cluster where it below.
[00051) COMPARING MODELS BY BAYESIAN INFORMATION CRITERION (BIC) [00052] Suppose the parameters have been estimated for models with different number of components, the "best" . model is selected according to the Bayesian Information Criterion (BIC). The BIC score is defined as), - (1/2) * v' log{n), where X is the value of log likelihood function using the estimated parameters, v is the number of independent parameters of the model, and n is the number of observation. The selected model is that with the highest BIC score.
[00053]For example, if there are two BIC s corresponding to a new model and an old model,they are named BIC new and B1Cw. BJCõ (1/2) * V,,* log{N), and BICw = X (.112)" Vow * log(N). The new model is accepted if SICõ BICew, That is, (1/2) * VneW* log(N) > 7lm - (1/2) * Võ4 * log(N), which is the same as X, -1b, (1/2) * log{N) * (vnew` V.W) [00054]' Since BIC is an approximation, it is not 100% accurate. A variable is added to control the BIC.. A variable, a, is therefore introduced, such that the model is selected if I.. - Xw > * (1/2) * log (N) * (vdEw V. In theory, the value of a is 1, but by changing the value of a, the model selected can be controlled For example, if a is set to be relatively small, then there is a high probability that the new model will be selected.
If a is set to be relatively large, then there is a high probability that the old model will be selected.
[00055]. Standardize the Data Set [00056) One problem with the data set is that each variable may have different shape in terms of the student t-distribution. 'T'herefore, each variable must be 'standardized order to let it follow a standard student t-distribution. The steps for such standardization are as follows:
[00057] Consider all.data points coming from one cluster. Set initial value of 1 to be the mean of the data points, and set the initial diagonal entries of E to be the variances of each variable, E0005$1 Find' the maximum likelihood estimates for 1,6 and Em, Create a vector with the diagcmal entries of E. and call it varm [000591 For a given data point xn, standardize it by using {xõ . õI/ varm E00060] METHOD STEPS
[000611 Using these statistical methods explained above, the method steps according to aspects ' of the present invention is now constructed.. If the vehicles have a different number of clusters, they can not be in the- same cluster. Hence, the data set can be classified into groups according to the number 'Of the -axles. Each group can therefore be partitioned into small groups by grouping vehicles with the saute axle pattern {s, d, t, or q) together.
Then the method steps are run inside each small group 13.
1) Standardize said data in setsi 2) When said data is standardized in sets, start with k 1;
3) Set the initial value of p , to be the mean of the data set.
4) Set the initial diagonal entries of Zk to be the variances of each variable.
S) Set P(K) =I .
6) Run clustering with the EM algorithm in this cluster.
7) Obtain the new values for 11k , Zk= P(k) and the probability matrix P (k I
x,)
[0068] Another aim of the invention is to provide better methodology for cluster analysis with application to the vehicle detection problem.
STATEMENT OF INVENTION
(0009] One aspect of the present invention provides a method of determining and recognizing -the' types of vehicles passing a checkpoint by collecting vehicle data (e.g., vehicle length, distances between axles, weights on axles) and using that data to determine and recognize vehicle types, particularly in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks, the method comprising: uploading a computer program into a CPU, the computer program comprising an EM algorithm as particularly described in the specification herein, the algorithm including data representations, of essential characteristics vehicles as they drive past the checkpoint; entering such measured characteristics vehicles as they travel past the checkpoint into that CPU; and deriving an output from that CPU, and thereby determining and recognizing the types of vehicles passing the checkpoint, particularly in high volume traffic, for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks.
[00101 Another aspect of the present invention provides an apparatus comprising the combination of, a CPU: and a computer program which has been uploaded into said CPU, the computer program comprising an EM algorithm as particularly described in the specification herein, the algorithm including data representations of essential characteristics of vehicles.
[00'11) It has been found according to aspects of the present invention, that there are correlations between different variables. This invention proposes to avoid the problem which arises by using the Euclidean distance, since data may be assigned to the wrong centrolds. The present invention seeks to overcome this problem by replacing the Euclidean distance with the Mahalanobis distance-(00121 The following description provides examples of methods of aspects of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS
In the accompanying drawings, [0013.) Fig 1 is .a graph showing some data points derived from traffic data collection systems which have collected data (e. g.., vehicle length, distances between axles, weights on axles, etc.), which data may be used to determine and recognize vehicle types in high volume traffic, but in which the cluster points are incorrectly clustered [0014] Fig 2 is a graph showing some data points derived from traffic data collection systems which have collected data (a. g.., vehicle length, distances between mules, weights on axles, etc.), which data may be used to determine and recognize Vehicle types in high volume traffic, but in which the cluster points are correctly clustered [0015] Fig 3 is a graph showing. some data points derived from traffic data collection systems which have collected data, (e. S.., vehicle length, distances between ,axles, weights on axles, etc.), which data may be used to deter-mine and recognize vehicle types in high volume traffic, where the X-means is used to cluster these points, using Euclidean distances, which is not correct.
[0016] one explanation of the results plotted in Figures I, 2 and 3 is because the X-means algorithin'does not permit returning back to re-cluster the data, since it runs a local K-means for each pair of "children". The K means is local in that the "children"
are fighting each other for the points in the "parent's"region and in no others. All the points from the other regions are ignored.
[0017] This problem of local mode can be overcome, according to broad aspects of the present invention, by merging two regions which are close to each other after the X-means algorithm is run. If the model after merging has a higher BIC score than the model before merging, these regions will be merged. Otherwise, the original model is kept.
ASSUMPTIONS
(00IS].In the method of determining and recognizing the types of vehicles passing a checkpoint by collecting vehicle data (e ,g., vehicle length, distances between axles, weights on axles) according to aspects of the present invention, by plotting graphs for all the variables (axle spacing, weights, front bumping spacing, and rear bumper spacing) in the data set, each variable forms a pattern which is similar to a "Student's" t. distribution. Therefore, it will be assumed that each variable in the data set comes from a "Student's" t-distribution Since all the variables for a given data point must' be considered, it will be assumed that each data point forms a multivariate "Student" distribution. In statistics, a "multivariate Student distribution" is a multivariate generalization of the Students t-distribution.
[40l9f The Present invention will be further described by reference to method steps to be carried out [ 00020] FINDING PARAMETER VALUES WITH EM ALGORITHM METHOD STEPS
[00021] The setup for the method steps is that given "N" data points in a V-dimensional space, it is desired to find a set of "K" "multivariate Student's t-distribution" that best represents the distribution of all the data points. Without being bound by theory, it is believed that the given data set is an N * V matrix, where N stands for the number of data points and V stands for the number of variables for each data point.
[OOO22] DEFT TION OF TERMS
N = number of data points.
V number of variables.
K number of dusters.
k = the mean for kill cluster, each a vector of length V.
E k:- the covariance matrices for kth duster, each of size V* V.
X11.= the nth data point, which is a vector with length V.
S
P(k ! xa:) - the probability that xn Comes from cluster k.
p(k) = : the probability that a data point chosen randomly comes from cluster k.
P(xn) = the probability of finding a data point at position xn = the value of log likelihood of the estimated parameter set.
(00023] For simplicity, it is assumed that k is a diagonal matrix, i.e., a matrix whose non diagonal entries are all 0, and where the diagonal entries are the variances for each variable.
[00024) Three statistical methods are used for the method steps, which are are carried out to find pararncter values with the EM algorithm, splitting clusters using Principle Component Analysis (PCA), and comparing models by Bayesian Information Criterion (BIC).
[00025] X is the key to tktis method. We it is not desired to be limited by any particular theory, it is believed that it is necessary to find the best values for the parameters by maximizing the value of I X. This method maximizes the posterior probability of the parameters if the specific priors are given.
[00026) The -steps are described as follows [00027] Step I
[00028] Set the=starting vahies forthe pks, Ek s, P (k). The method to obtain these values is by means of splitting clusters using PCA as follows:
[00029) The setup for this method is that, given some data points in one cluster (mother cluster),it is necessary to split these data points into two clusters (children clusters), using PCA.
PCA is mathestiatjcally deAned'as "an orthogonal linear transformation that transforms the data to anew coordinate system such that the greatest variance by any projection of the data comes to lie on the fast coordinate (called the first principal component), the second greatest 'variance on the second coordinate", and so on. The data set comprises an N * matrix. PCA
is .now performed an this data matrix. The standard deviations are now calculated of the principal components, namely, the square roots of the eigenvalues of the covariance matrix. The matrix is now calculated far variable loadings, namely, a matrix whose columns contain the eigenvectors.
In R, the is a built-in function called "preomp" which helps the calculations.
(00030] The terms used are defined as follows:
std = a vector contains the square roots of the eigenvalues of the covariance matrix.
Rotation = a matrix whose columns contain the eigenvectors Range: how far the data is to be split; the.value of range is usually between 1.5 and K = the mean for the mother cluster, each a vector of length V
E = the covariance matrix for the mother cluster, each of size 'P(M) = the probability that a data point chosen randomly comes from the ritother cluster (0003 i] Since the first principal component is the most important component, two vectors are created with length V from the first principal component. Two vectors are created since it is desired to to split the data into two clusters. The first element in. the first vector is the value of range (plus range) , and the other elements are all zero. The first element in the second vector is the value of -range (minus range) . and the other elements are all zero. The first vector is V 1, and the second vector is V 2. Consider V 1 , V2, and std to be matrices with one column. After the splitting is done, two meads are provided for two children clusters. The mean for the first children cluster is placid the mean for the second cluldren cluster }-2.The calculation for gland is as follows:
Pl li+ rotation W% (V 1 * Std) p2= 1,E+ rotation %* %o (V2 std) [000321 Here, k*k and "%.*%" different operations.
(00033]For example, a d a+d a d g j a*j +d*h+g*1 b*e -b*e but beh%*% k b*j +e*k+h*1 c f c*f c f i I c+j +f*k+l+l [0034] The covariance matrices for two children clusters would be the same as the mother cluster, and the probability that a data point chosen randomly comes from the children clusters would be half ofP(m).
[000351 In stmunary, for the first children cluster, mean = l , covariance matrix = E, and probability that a data point chosen randomly comes from this cluster = (I
/2)*P{m); for the second children cluster, mean 112, covariance matrix = E, and probability that a data point chosen randomly comes from this cluster = (1/2)*P(m).
[00030] There is one limitation about PCA. If PCA is performed on a given data matrix, PCA
requires 'the number of data points'to be larger than the number of variables.
If the number of data points is smaller than the number of variables, PCA will do nothing on the data matrix, and splitting will not happen [000361 Step 2-[0003.71 Given the values for the k's, F..k s, and P (k), and the data, of P
(xn 1 k , Ek) can be calculated. we assume all the variables in the data set form a multivariate Student's distribution, P (xn - k, Mk) is the multivariate Student's density, that is, rj td_'~
P (xul Pe. fir) {dr r (doh (detE) z I t (s) * (xa ^ Ne)F . E-A . (xn - pk}
where df w the degree of freedom, p T the number of variables, and detE = the determinant for [00037] One important thing about P (xn l k , Ek) is that the values of P (xn 1 Ak , FY often be very small as to underflow to zero. Thereforc,it is necessary to work with the logarithm ofP (xn 1 k Lk) instead of P(zn 1 k , Ek), that is 10$ P (X.1 uk= F-k) logr(df+p) - logr (d2) - log (rc)gz - log(df)Pz - log(det'E)z -iogl+\_) (xõ_1i*)Ts E r(,xx-Pk)J
[00038]After the value'of P (xn Ilk , Ek) is obtained, it becomes possible to calculate the value of P (xc) splitting P (xn) into its contribution from each of the K
multivariate Student's t-distributions, that is, P (xii) P (xn I Pk, Ek) P(k) k [00039] One problem may rise for P(xn), where it becomes necessary to calculate the sum of quantities. Some of these quantities may be so small that they underflow to zero. According to an aspect of the present invention, it has been found that one possible way to fix this problem is construct the quantities from their Logarithms. That is, store P (xn 1 k , F.k) P(i) in log P (xn l k , Zk), and let m, - max log (P (xn 10k , )-k)P(i)0..., log P(k))). Then the logarithm of the sum is computed as follows:
IagP (x,1) a log(m ,,) + log(l exp(log (P(; l1 L. Ef)F(t)) - mraa)) 1000401 Using the values of P ,(xn I k , Zk) and P (xn), the value of P(kl xn) and 3.:
P(klx,) = P(XRII~~=EK)P(k) P (xõ) A = tog (fl P(xõ )) _ X logP (x') 100041 ] Since the values of log P (x, ' I l k Ek) and log P (xn) can be computed in order to overcome the problem of underflow, it is possible to write P(k 1 xn) in terms of log P
(xn l k , Ek) and I. P (xn), p(klxx) = exp (IogP(xnl Lk.Er) + iogP(k) - logP(x,,)) [00042]By calculating P (ac I xn) for all values of k's and xõ's, it is now possible to obtain all.of the value's P (k I x,1}'s, and it now becomes possible to write P (k I xa)'s as a probability matrix of size N * K. Each row = one data point, and each column =
one cluster. Each element in the matrix = the value of P (k l xn), that is, the probability that a given data point comes from a specific cluster k. In the language of the EM
algorithm, this is called an expectation step or an E-step.
[00043] Step 3:
[00] Using P (k I Qs from step 2, the values of maximum likelihood estimates for k's is, Ek s and P(k)'s and for all values of k, can be calculated, that is , the values of k's , Zk's and Pik)'s that maximize the log likelihood function X. The maximum likelihood estimate for P(k) is easy to obtain-P(k) p (klx=,) 1000451 The process to calculate the maximum likelihood estimates for id's , rk's is very complex. For a given cluster k (k = t, 2, 3 ...K , ), the log likelihood function needed to maximize is as follows-A log (P (x l k=Ek)) * P(klx.) [00046j'11 is now necessary to find the values of pk and F..k'sfor k =1, 2, 3 ...K that maximize the above function.
[00047) Most of the programming languages have the build-in functions to calculate the values of the parameters that maximize a given function. For example, in R, we can use a build-in fimction called "DIM" can be used to calculate the maximum likelihood estimates for -k'S 'S. the language of the EM steps is called a maximization step or M-step.
[00045] Step 4;
[000491 Using the maxirnuri likelihood estimates for ttk's , F..k's and P(k}'s as the new pi's, F.k's and F(k}'s, repeat Step 2 and Step 3 until the value of X no longer changes.
(00050] After the clustering process, the final values for gl s , E, s and P(k)'s have been obtained for all values of k. A probability matrix whose entries, are the final values of F (k I Qs have also been obtained Given a data point,the corresponding row in the probability matrix for this data point can be found Then, it, is possible to determine which cluster most likely comes from by looking at the values of P
(k I ys from this row. The column index which produces the largest value of P (k I xn) is the cluster where it below.
[00051) COMPARING MODELS BY BAYESIAN INFORMATION CRITERION (BIC) [00052] Suppose the parameters have been estimated for models with different number of components, the "best" . model is selected according to the Bayesian Information Criterion (BIC). The BIC score is defined as), - (1/2) * v' log{n), where X is the value of log likelihood function using the estimated parameters, v is the number of independent parameters of the model, and n is the number of observation. The selected model is that with the highest BIC score.
[00053]For example, if there are two BIC s corresponding to a new model and an old model,they are named BIC new and B1Cw. BJCõ (1/2) * V,,* log{N), and BICw = X (.112)" Vow * log(N). The new model is accepted if SICõ BICew, That is, (1/2) * VneW* log(N) > 7lm - (1/2) * Võ4 * log(N), which is the same as X, -1b, (1/2) * log{N) * (vnew` V.W) [00054]' Since BIC is an approximation, it is not 100% accurate. A variable is added to control the BIC.. A variable, a, is therefore introduced, such that the model is selected if I.. - Xw > * (1/2) * log (N) * (vdEw V. In theory, the value of a is 1, but by changing the value of a, the model selected can be controlled For example, if a is set to be relatively small, then there is a high probability that the new model will be selected.
If a is set to be relatively large, then there is a high probability that the old model will be selected.
[00055]. Standardize the Data Set [00056) One problem with the data set is that each variable may have different shape in terms of the student t-distribution. 'T'herefore, each variable must be 'standardized order to let it follow a standard student t-distribution. The steps for such standardization are as follows:
[00057] Consider all.data points coming from one cluster. Set initial value of 1 to be the mean of the data points, and set the initial diagonal entries of E to be the variances of each variable, E0005$1 Find' the maximum likelihood estimates for 1,6 and Em, Create a vector with the diagcmal entries of E. and call it varm [000591 For a given data point xn, standardize it by using {xõ . õI/ varm E00060] METHOD STEPS
[000611 Using these statistical methods explained above, the method steps according to aspects ' of the present invention is now constructed.. If the vehicles have a different number of clusters, they can not be in the- same cluster. Hence, the data set can be classified into groups according to the number 'Of the -axles. Each group can therefore be partitioned into small groups by grouping vehicles with the saute axle pattern {s, d, t, or q) together.
Then the method steps are run inside each small group 13.
1) Standardize said data in setsi 2) When said data is standardized in sets, start with k 1;
3) Set the initial value of p , to be the mean of the data set.
4) Set the initial diagonal entries of Zk to be the variances of each variable.
S) Set P(K) =I .
6) Run clustering with the EM algorithm in this cluster.
7) Obtain the new values for 11k , Zk= P(k) and the probability matrix P (k I
x,)
8) Define the BIC for this model as B=lCold= - (112) = V,,, = log{N)
9) Set k_p,*v . k and Repeat the following steps until k~,e k a) Set k~, = Ic and a new variable called trace =1.
b) Repeat the following steps until trace me _ka..
(i) Split the cluster at position trace into two clusters using PCA
(ii) Select data points to perform PCA., from the data points that are most likely come from cluster trace by checking the values in the probability matrix P (k I x ) (iii) Run clustering with the EM algorithm for this new model, (iv) Obtain pk's , E:k's and P(k}'s and the probability matrix P (k I
xõ )'s, and for the new model, 1'4 (v) ,Define the BIC for this new model as BICnew; -X... - (112) =
Vnew, log(N) (vi) If > a' (112) = (N) ' (Vnew- Vw), then replace the old model with the new model obtained in step (iii).
(vii) Set K-+ 1 (uiii)lf' " - X,,, is not > a= (4) * (N) = (vnew Vw), then keep the ot'iginal model.
(ix) Trace = trace + I
11) Finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks, wherein, in said steps, N - number of data points.
V - number of variables.
K - number ofdu sters.
ilk = the meat, for kill cluster, each a vector of length V, fa k:- the covariance matrices for kth duster, each of size V ` V.
xn.= the nth data point, which is a vector with length V.
P(k l xn:) - the probability that xn comes from cluster k.
p(k) -: the probability that a data point chosen randomly comes from cluster k.
P(xn) - the probability of finding a data point at position xn X - the value of log likelihood of the estimated parameter set.
PCA Principal Component Analysis BIC = Bayesian Information Criterion C00062]The clustering result is obtained by checking the values ,in the final probability matrix.
[00063] Using the final model, any data points may be clustered , For example, if some data points are. given, they can be assigned to their corresponding clusters.
Using Step 2 to 8 as defined above with the EM algorithm method step a probability matrix can be obtained whose entries arc the values of P (k l xn) for all values of cluster k's and x.'sw From these values it is possible to determine which cluster that each data point most likely comes from by checking the values in the probability matrix.
[000641 If there is a vehicle that seldom appears, and it is desired to cluster it into a single cluster once it appears, this can be accomplished by adding a new cluster to the final model. The value of la for the new cluster is the same as the variable values for this vehicle. The covariance matrix E is a diagonal matrix. The diagonal entries of E is set to be very small numbers, namely the variances for each variable are small numbers. The values of P(k) for this cluster are, set to be a small number since this vehicle is very rare to appear.
b) Repeat the following steps until trace me _ka..
(i) Split the cluster at position trace into two clusters using PCA
(ii) Select data points to perform PCA., from the data points that are most likely come from cluster trace by checking the values in the probability matrix P (k I x ) (iii) Run clustering with the EM algorithm for this new model, (iv) Obtain pk's , E:k's and P(k}'s and the probability matrix P (k I
xõ )'s, and for the new model, 1'4 (v) ,Define the BIC for this new model as BICnew; -X... - (112) =
Vnew, log(N) (vi) If > a' (112) = (N) ' (Vnew- Vw), then replace the old model with the new model obtained in step (iii).
(vii) Set K-+ 1 (uiii)lf' " - X,,, is not > a= (4) * (N) = (vnew Vw), then keep the ot'iginal model.
(ix) Trace = trace + I
11) Finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks, wherein, in said steps, N - number of data points.
V - number of variables.
K - number ofdu sters.
ilk = the meat, for kill cluster, each a vector of length V, fa k:- the covariance matrices for kth duster, each of size V ` V.
xn.= the nth data point, which is a vector with length V.
P(k l xn:) - the probability that xn comes from cluster k.
p(k) -: the probability that a data point chosen randomly comes from cluster k.
P(xn) - the probability of finding a data point at position xn X - the value of log likelihood of the estimated parameter set.
PCA Principal Component Analysis BIC = Bayesian Information Criterion C00062]The clustering result is obtained by checking the values ,in the final probability matrix.
[00063] Using the final model, any data points may be clustered , For example, if some data points are. given, they can be assigned to their corresponding clusters.
Using Step 2 to 8 as defined above with the EM algorithm method step a probability matrix can be obtained whose entries arc the values of P (k l xn) for all values of cluster k's and x.'sw From these values it is possible to determine which cluster that each data point most likely comes from by checking the values in the probability matrix.
[000641 If there is a vehicle that seldom appears, and it is desired to cluster it into a single cluster once it appears, this can be accomplished by adding a new cluster to the final model. The value of la for the new cluster is the same as the variable values for this vehicle. The covariance matrix E is a diagonal matrix. The diagonal entries of E is set to be very small numbers, namely the variances for each variable are small numbers. The values of P(k) for this cluster are, set to be a small number since this vehicle is very rare to appear.
Claims (11)
1. A method for determining and recognizing types of vehicles passing a check point, which comprises:
up-loading an EM algorithm into a CPU;
collecting vehicle data as vehicles drive past a check point;
entering said data into said CPU said data being representative of essential characteristics of vehicles;
processing said data by said EM algorithm to produce an output model of the traffic volumes of the various types of vehicles; and utilizing said output model to forecast future road maintenance costs and/or to plan and design future road networks, wherein said EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of pk to be the mean of the data set;
4) set the initial diagonal entries of a to be the variances of each variable;
5) set P(K)=-1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold.cndot.log(N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew¨.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot. (1/2).cndot.
(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
up-loading an EM algorithm into a CPU;
collecting vehicle data as vehicles drive past a check point;
entering said data into said CPU said data being representative of essential characteristics of vehicles;
processing said data by said EM algorithm to produce an output model of the traffic volumes of the various types of vehicles; and utilizing said output model to forecast future road maintenance costs and/or to plan and design future road networks, wherein said EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of pk to be the mean of the data set;
4) set the initial diagonal entries of a to be the variances of each variable;
5) set P(K)=-1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold.cndot.log(N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew¨.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot. (1/2).cndot.
(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
2. The method of claim 1, wherein said vehicle data comprises length of said vehicle, distance between axles of said vehicle, and weights on said axles of said vehicle.
3. The method of claim 1, including the additional step of obtaining the clustering results by checking the values in the final probability matrix.
4. The method of claim 1 including the additional step of using the final model, to cluster many data points where, if some data points are given, they can be assigned to their corresponding clusters, and obtaining a probability matrix whose entries are the values of P (k 1 x n) for all values of cluster k's and x n's, and from these values determining which cluster that each data point most likely comes from by checking the values in the probability matrix.
5. The method of claim 1, for a vehicle that seldom appears, and it is desired to cluster it into a single cluster once it appears, by adding a new cluster to the final model, where the value of µ for the new cluster is the same as the variable values for this vehicle, where the covariance matrix .SIGMA. is a diagonal matrix, and setting diagonal entries of .SIGMA. to be very small numbers, where the variances for each variable are small numbers so that the values of P(k) for this cluster are set to be a small number since this vehicle is very rare to appear.
6. A method of determining and recognizing the types of vehicles passing a checkpoint which comprises the steps of:
uploading a computer program into a CPU, said computer program comprising an EM
algorithm said EM algorithm including data representations of essential characteristics of vehicles collecting vehicle data as said vehicles drive past a checkpoint to determine and recognize vehicle types for monitoring traffic volumes of various types of vehicles, entering said data into said CPU; and deriving an output from said CPU, thereby monitoring traffic volumes of various types of vehicles for forecasting future road maintenance costs and planning and design of future road networks, wherein the EM algorithm algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold.cndot.log (N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=-.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks;
wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
uploading a computer program into a CPU, said computer program comprising an EM
algorithm said EM algorithm including data representations of essential characteristics of vehicles collecting vehicle data as said vehicles drive past a checkpoint to determine and recognize vehicle types for monitoring traffic volumes of various types of vehicles, entering said data into said CPU; and deriving an output from said CPU, thereby monitoring traffic volumes of various types of vehicles for forecasting future road maintenance costs and planning and design of future road networks, wherein the EM algorithm algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold.cndot.log (N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=-.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks;
wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
7. An apparatus comprising the combination of:
a CPU; and a computer program which has been uploaded into said CPU, said computer program comprising an EM algorithm, said EM algorithm including data representations of essential characteristics of vehicles, wherein the EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold-log(N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_wev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=¨.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
a CPU; and a computer program which has been uploaded into said CPU, said computer program comprising an EM algorithm, said EM algorithm including data representations of essential characteristics of vehicles, wherein the EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=¨(1/2).cndot.Vold-log(N);
9) set k_prev=.k; and 10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_wev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=¨.lambda.new¨(1/2)*Vnew*log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K=+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(1/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace=trace+1; and 11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn.=the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
8. The apparatus of claim 7 wherein said vehicle data comprises length of said vehicle, distance between axles of said vehicle and weights on said axles of said vehicle.
9. An apparatus for determining and recognizing types of vehicles passing a check point, which comprises:
a CPU;
an EM algorithm uploaded into said CPU;
structure operatively associated with said CPU for collecting vehicle data as vehicles drive past said check point;
means, operatively associated with said CPU for entering said data into said CPU said data being representative of essential characteristics of vehicles;
means for processing said data by said EM algorithm to produce an output model of the traffic volumes of the various types of vehicles; and means for utilizing said output model to forecast future road maintenance costs and/or to plan and design future road networks, wherein the EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k a to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=-(1/2).cndot.Vold.cndot.log (N);
9) set k_prev=.cndot.k; and
a CPU;
an EM algorithm uploaded into said CPU;
structure operatively associated with said CPU for collecting vehicle data as vehicles drive past said check point;
means, operatively associated with said CPU for entering said data into said CPU said data being representative of essential characteristics of vehicles;
means for processing said data by said EM algorithm to produce an output model of the traffic volumes of the various types of vehicles; and means for utilizing said output model to forecast future road maintenance costs and/or to plan and design future road networks, wherein the EM algorithm is specially adapted to carry out the following steps:
1) standardize said data in sets;
2) when said data is standardized in sets, start with k 1;
3) set the initial value of µk to be the mean of the data set;
4) set the initial diagonal entries of .SIGMA.k a to be the variances of each variable;
5) set P(K)=1;
6) run clustering with the EM algorithm in this cluster;
7) obtain the new values for µk, .SIGMA.k, (Pk) and the probability matrix P (k 1 xn);
8) define the BIC for this model as BICold=-(1/2).cndot.Vold.cndot.log (N);
9) set k_prev=.cndot.k; and
10) repeat the following steps until k_prev=k;
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=¨.lambda.new¨(1/2).cndot.Vnew.cndot.log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew--Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace trace+1; and
a) set k_prev=k and a new variable called trace=1;
b) repeat the following steps until trace=k_prev;
(i) split the cluster at position trace into two clusters using PCA;
(ii) select data points to perform PCA from the data points that are most likely to come from cluster trace by checking the values in the probability marix P (k 1 xn);
(iii) run clustering with the EM algorithm for this new model;
(iv) obtain µk's, .SIGMA.k's and (Pk)'s and the probability matrix P (k 1 xn)'s, and for the new model;
(v) define the BIC for this new model as BICnew=¨.lambda.new¨(1/2).cndot.Vnew.cndot.log(N);
(vi) if .lambda.new¨.lambda.old>a.cndot.(1/2).cndot.(N).cndot.(vnew--Vold), then replace the old model with the new model obtained in step (iii);
(vii) set K+1;
(viii) if .lambda.new¨.lambda.old is not >a.cndot.(/2).cndot.(N).cndot.(vnew¨Vold), then keep the original model;
(ix) trace trace+1; and
11) finally report the final model;
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn. =the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
thereby determining and recognizing the types of vehicles passing the checkpoint to determine and recognize vehicle types in high volume traffic for monitoring traffic volumes of various types of vehicles, forecasting future road maintenance costs and planning and design of future road networks; wherein in said steps:
N=number of data points;
V=number of variables;
K=number of clusters;
µk=the mean for kill cluster, each a vector of length V;
f ° k:=the covariance matrices for kth cluster, each of size V*V;
xn. =the nth data point, which is a vector with length V;
P(k ! xn:)=the probability that xn comes from cluster k;
p(k)=: the probability that a data point chosen randomly comes from cluster k;
P(xn)=the probability of finding a data point at position xn;
.lambda.=the value of log likelihood of the estimated parameter set;
PCA=Principal Component Analysis; and BIC=Bayesian Information Criterion.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15486609P | 2009-02-24 | 2009-02-24 | |
US61/154,886 | 2009-02-24 |
Publications (1)
Publication Number | Publication Date |
---|---|
CA2694451A1 true CA2694451A1 (en) | 2010-08-24 |
Family
ID=42630562
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA2694451A Abandoned CA2694451A1 (en) | 2009-02-24 | 2010-02-24 | Cluster and discriminant analysis for vehicles detection |
Country Status (2)
Country | Link |
---|---|
US (5) | US20100214282A1 (en) |
CA (1) | CA2694451A1 (en) |
Families Citing this family (57)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100214282A1 (en) | 2009-02-24 | 2010-08-26 | Dolby Laboratories Licensing Corporation | Apparatus for providing light source modulation in dual modulator displays |
US10002571B1 (en) * | 2010-02-26 | 2018-06-19 | Zulch Laboratories, Inc. | Liquid crystal display incorporating color-changing backlight |
US8735791B2 (en) | 2010-07-13 | 2014-05-27 | Svv Technology Innovations, Inc. | Light harvesting system employing microstructures for efficient light trapping |
CN103262144B (en) * | 2010-12-17 | 2016-08-17 | 杜比实验室特许公司 | Quantum dot for display floater |
US9373178B2 (en) | 2011-08-24 | 2016-06-21 | Dolby Laboratories Licensing Corporation | High dynamic range displays having wide color gamut and energy efficiency |
US9082349B2 (en) * | 2011-08-30 | 2015-07-14 | Sharp Laboratories Of America, Inc. | Multi-primary display with active backlight |
US9197881B2 (en) * | 2011-09-07 | 2015-11-24 | Intel Corporation | System and method for projection and binarization of coded light patterns |
US9324250B2 (en) | 2011-09-09 | 2016-04-26 | Dolby Laboratories Licensing Corporation | High dynamic range displays comprising MEMS/IMOD components |
US9097826B2 (en) | 2011-10-08 | 2015-08-04 | Svv Technology Innovations, Inc. | Collimating illumination systems employing a waveguide |
WO2013078249A1 (en) | 2011-11-22 | 2013-05-30 | Qd Vision Inc. | Method of making quantum dots |
US10008631B2 (en) | 2011-11-22 | 2018-06-26 | Samsung Electronics Co., Ltd. | Coated semiconductor nanocrystals and products including same |
US9747866B2 (en) | 2011-11-22 | 2017-08-29 | Dolby Laboratories Licensing Corporation | Optimizing light output profile for dual-modulation display performance |
WO2013078242A1 (en) | 2011-11-22 | 2013-05-30 | Qd Vision, Inc. | Methods for coating semiconductor nanocrystals |
WO2013078247A1 (en) | 2011-11-22 | 2013-05-30 | Qd Vision, Inc. | Methods of coating semiconductor nanocrystals, semiconductor nanocrystals, and products including same |
WO2013078245A1 (en) | 2011-11-22 | 2013-05-30 | Qd Vision, Inc. | Method of making quantum dots |
KR101960469B1 (en) | 2012-02-05 | 2019-03-20 | 삼성전자주식회사 | Semiconductor nanocrystals, methods for making same, compositions, and products |
EP2862162B1 (en) | 2012-06-15 | 2020-03-18 | Dolby Laboratories Licensing Corporation | Systems and methods for controlling dual modulation displays |
KR102118309B1 (en) | 2012-09-19 | 2020-06-03 | 돌비 레버러토리즈 라이쎈싱 코오포레이션 | Quantum dot/remote phosphor display system improvements |
US20140204039A1 (en) * | 2013-01-22 | 2014-07-24 | Adobe Systems Incorporated | Compositing display |
KR20140101200A (en) * | 2013-02-08 | 2014-08-19 | 삼성전자주식회사 | Display device |
BR112015020571B1 (en) * | 2013-03-08 | 2022-04-12 | Dolby Laboratories Licensing Corporation | Method for triggering a local dimming monitor, computer readable non-transient storage medium and device |
US9617472B2 (en) | 2013-03-15 | 2017-04-11 | Samsung Electronics Co., Ltd. | Semiconductor nanocrystals, a method for coating semiconductor nanocrystals, and products including same |
US9224323B2 (en) | 2013-05-06 | 2015-12-29 | Dolby Laboratories Licensing Corporation | Systems and methods for increasing spatial or temporal resolution for dual modulated display systems |
ES2768699T3 (en) * | 2013-07-30 | 2020-06-23 | Dolby Laboratories Licensing Corp | Projector screen systems that have non-mechanical mirror beam direction |
KR20150037368A (en) | 2013-09-30 | 2015-04-08 | 삼성전자주식회사 | Modulator array, Moduating device and Medical imaging apparatus comprising the same |
EP3080799A4 (en) * | 2013-12-10 | 2017-12-06 | Dolby Laboratories Licensing Corporation | Laser diode driven lcd quantum dot hybrid displays |
EP3123240A2 (en) | 2014-03-26 | 2017-02-01 | Dolby Laboratories Licensing Corp. | Global light compensation in a variety of displays |
JP6236188B2 (en) | 2014-08-21 | 2017-11-22 | ドルビー ラボラトリーズ ライセンシング コーポレイション | Dual modulation technology with light conversion |
US20170061894A1 (en) * | 2015-08-26 | 2017-03-02 | Canon Kabushiki Kaisha | Image display apparatus |
CN113406849B (en) * | 2017-05-17 | 2022-04-15 | 深圳光峰科技股份有限公司 | Excitation light intensity control method |
KR102496683B1 (en) | 2017-10-11 | 2023-02-07 | 삼성디스플레이 주식회사 | Display panel and display device comprising the display panel |
US20190172415A1 (en) * | 2017-12-01 | 2019-06-06 | Dennis Willard Davis | Remote Color Matching Process and System |
WO2019117913A1 (en) | 2017-12-14 | 2019-06-20 | Hewlett-Packard Development Company, L.P. | Displays with phosphorescent components |
CN109003568A (en) * | 2018-09-13 | 2018-12-14 | 天长市辉盛电子有限公司 | LED display point-to-point correction system and method |
US11315467B1 (en) | 2018-10-25 | 2022-04-26 | Baylor University | System and method for a multi-primary wide gamut color system |
US11587491B1 (en) | 2018-10-25 | 2023-02-21 | Baylor University | System and method for a multi-primary wide gamut color system |
US11189210B2 (en) | 2018-10-25 | 2021-11-30 | Baylor University | System and method for a multi-primary wide gamut color system |
US11069280B2 (en) | 2018-10-25 | 2021-07-20 | Baylor University | System and method for a multi-primary wide gamut color system |
US11062638B2 (en) | 2018-10-25 | 2021-07-13 | Baylor University | System and method for a multi-primary wide gamut color system |
US11373575B2 (en) | 2018-10-25 | 2022-06-28 | Baylor University | System and method for a multi-primary wide gamut color system |
US11289003B2 (en) | 2018-10-25 | 2022-03-29 | Baylor University | System and method for a multi-primary wide gamut color system |
US10607527B1 (en) | 2018-10-25 | 2020-03-31 | Baylor University | System and method for a six-primary wide gamut color system |
US11403987B2 (en) | 2018-10-25 | 2022-08-02 | Baylor University | System and method for a multi-primary wide gamut color system |
US11043157B2 (en) | 2018-10-25 | 2021-06-22 | Baylor University | System and method for a six-primary wide gamut color system |
US11289000B2 (en) | 2018-10-25 | 2022-03-29 | Baylor University | System and method for a multi-primary wide gamut color system |
US11488510B2 (en) | 2018-10-25 | 2022-11-01 | Baylor University | System and method for a multi-primary wide gamut color system |
US11069279B2 (en) | 2018-10-25 | 2021-07-20 | Baylor University | System and method for a multi-primary wide gamut color system |
US11475819B2 (en) | 2018-10-25 | 2022-10-18 | Baylor University | System and method for a multi-primary wide gamut color system |
US11341890B2 (en) | 2018-10-25 | 2022-05-24 | Baylor University | System and method for a multi-primary wide gamut color system |
US11037481B1 (en) | 2018-10-25 | 2021-06-15 | Baylor University | System and method for a multi-primary wide gamut color system |
US10950162B2 (en) | 2018-10-25 | 2021-03-16 | Baylor University | System and method for a six-primary wide gamut color system |
US10997896B2 (en) | 2018-10-25 | 2021-05-04 | Baylor University | System and method for a six-primary wide gamut color system |
US11030934B2 (en) | 2018-10-25 | 2021-06-08 | Baylor University | System and method for a multi-primary wide gamut color system |
US11532261B1 (en) | 2018-10-25 | 2022-12-20 | Baylor University | System and method for a multi-primary wide gamut color system |
US11410593B2 (en) | 2018-10-25 | 2022-08-09 | Baylor University | System and method for a multi-primary wide gamut color system |
US10950161B2 (en) | 2018-10-25 | 2021-03-16 | Baylor University | System and method for a six-primary wide gamut color system |
KR102608147B1 (en) | 2018-12-05 | 2023-12-01 | 삼성전자주식회사 | Display apparatus and driving method thereof |
Family Cites Families (350)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4316196A (en) * | 1977-03-10 | 1982-02-16 | Bell & Howell Corporation | Illumination and light gate utilization methods and apparatus |
US4170771A (en) * | 1978-03-28 | 1979-10-09 | The United States Of America As Represented By The Secretary Of The Army | Orthogonal active-passive array pair matrix display |
US4229095A (en) * | 1979-01-29 | 1980-10-21 | Eastman Kodak Company | Electro-optical color imaging apparatus |
US4364039A (en) * | 1980-07-25 | 1982-12-14 | Texas Instruments Incorporated | Stacked electro-optic display |
US4441791A (en) * | 1980-09-02 | 1984-04-10 | Texas Instruments Incorporated | Deformable mirror light modulator |
US4378568A (en) * | 1981-01-29 | 1983-03-29 | Eastman Kodak Company | Light valve imaging apparatus and method for providing gray scale |
US4374397A (en) * | 1981-06-01 | 1983-02-15 | Eastman Kodak Company | Light valve devices and electronic imaging/scan apparatus with locationally-interlaced optical addressing |
JPS5810481U (en) * | 1981-07-10 | 1983-01-22 | シャープ株式会社 | liquid crystal display device |
FR2536563B1 (en) * | 1982-11-23 | 1985-07-26 | Ssih Equipment Sa | LIGHT EMITTING ELEMENT WITH DISCHARGE TUBE FOR MATRIX DISPLAY BOARD |
JPS6054174A (en) | 1983-09-01 | 1985-03-28 | Seiko Instr & Electronics Ltd | Multiple battery |
JPS6054120A (en) | 1983-09-01 | 1985-03-28 | アルプス電気株式会社 | Method of producing membrane type pushbutton switch |
JPS6054174U (en) | 1983-09-20 | 1985-04-16 | 三洋電機株式会社 | Color liquid crystal display device |
JPS6054120U (en) | 1983-09-20 | 1985-04-16 | 三洋電機株式会社 | liquid crystal display device |
DE3581546D1 (en) | 1984-03-12 | 1991-03-07 | Matsushita Electric Ind Co Ltd | OPTICAL FILTER AND PRODUCTION METHOD. |
NL8401605A (en) * | 1984-05-18 | 1985-12-16 | Optische Ind De Oude Delft Nv | LIGHT BOX FOR GIVING A BACKGROUND LIGHT WITH BRIGHTNESS VALUES ADAPTED TO THE BLACK OF A LIGHT BOX FOR VIEWING. |
JPS6218593A (en) * | 1985-07-17 | 1987-01-27 | シャープ株式会社 | Data processor |
JPS62234133A (en) | 1986-04-04 | 1987-10-14 | Nec Corp | Flat display panel |
US4868668A (en) * | 1986-08-21 | 1989-09-19 | Electrohome Limited | System and method for image adjustment in an optical projection system |
DE3785813T2 (en) * | 1986-09-20 | 1993-11-11 | Emi Plc Thorn | Display device. |
US4726663A (en) * | 1986-11-14 | 1988-02-23 | Tektronix, Inc. | Switchable color filter with enhanced transmissivity |
US4801194A (en) * | 1987-09-23 | 1989-01-31 | Eastman Kodak Company | Multiplexed array exposing system having equi-angular scan exposure regions |
US4933754A (en) * | 1987-11-03 | 1990-06-12 | Ciba-Geigy Corporation | Method and apparatus for producing modified photographic prints |
US4987410A (en) * | 1988-01-25 | 1991-01-22 | Kaiser Aerospace & Electronics Corporation | Multiple image forming apparatus |
JPH01200232A (en) * | 1988-02-04 | 1989-08-11 | Sharp Corp | Ferroelectric liquid crystal display device |
JPH0278393A (en) | 1988-09-14 | 1990-03-19 | Hitachi Ltd | Stereoscopic color picture display device |
GB8823490D0 (en) * | 1988-10-06 | 1988-11-16 | Emi Plc Thorn | Method & apparatus for projecting scanned two/threedimensional modulated light pattern originating from light source |
US5050965A (en) * | 1989-09-01 | 1991-09-24 | In Focus Systems, Inc. | Color display using supertwisted nematic liquid crystal material |
JPH0341890A (en) | 1989-07-07 | 1991-02-22 | Pioneer Electron Corp | Beam index type color display device |
US5247366A (en) * | 1989-08-02 | 1993-09-21 | I Sight Ltd. | Color wide dynamic range camera |
JP2582644B2 (en) * | 1989-08-10 | 1997-02-19 | 富士写真フイルム株式会社 | Flat panel image display |
US4954789A (en) * | 1989-09-28 | 1990-09-04 | Texas Instruments Incorporated | Spatial light modulator |
JPH03198026A (en) | 1989-12-27 | 1991-08-29 | Hitachi Ltd | Liquid crystal display device, back light control system, and information processor |
JPH07121120B2 (en) | 1990-03-19 | 1995-12-20 | 日本ビクター株式会社 | Data compression device |
US5075789A (en) * | 1990-04-05 | 1991-12-24 | Raychem Corporation | Displays having improved contrast |
GB9008032D0 (en) | 1990-04-09 | 1990-06-06 | Rank Brimar Ltd | Video display systems |
GB9008031D0 (en) * | 1990-04-09 | 1990-06-06 | Rank Brimar Ltd | Projection systems |
FR2669744B1 (en) * | 1990-11-23 | 1994-03-25 | Thomson Csf | LIGHTING DEVICE AND APPLICATION TO A VISUALIZATION DEVICE. |
JPH04204591A (en) | 1990-11-30 | 1992-07-24 | Toshiba Corp | Projection type liquid crystal display device |
FR2664712B1 (en) * | 1991-10-30 | 1994-04-15 | Thomson Csf | OPTICAL MODULATION DEVICE WITH DEFORMABLE CELLS. |
US5359345A (en) * | 1992-08-05 | 1994-10-25 | Cree Research, Inc. | Shuttered and cycled light emitting diode display and method of producing the same |
US5724062A (en) * | 1992-08-05 | 1998-03-03 | Cree Research, Inc. | High resolution, high brightness light emitting diode display and method and producing the same |
US5461397A (en) * | 1992-10-08 | 1995-10-24 | Panocorp Display Systems | Display device with a light shutter front end unit and gas discharge back end unit |
DE69427860T2 (en) | 1993-02-03 | 2002-04-11 | Nitor San Jose | METHOD AND DEVICE FOR PROJECTING IMAGES |
GB2278480A (en) * | 1993-05-25 | 1994-11-30 | Sharp Kk | Optical apparatus |
CN1051379C (en) * | 1993-10-05 | 2000-04-12 | 梯尔技术公司 | Light source for back lighting |
US5440197A (en) * | 1993-10-05 | 1995-08-08 | Tir Technologies, Inc. | Backlighting apparatus for uniformly illuminating a display panel |
JPH07121120A (en) | 1993-10-25 | 1995-05-12 | Fujitsu Ltd | Plasma display unit |
US5748828A (en) * | 1993-11-10 | 1998-05-05 | Alliedsignal Inc. | Color separating backlight |
JP3213462B2 (en) * | 1993-11-25 | 2001-10-02 | 三洋電機株式会社 | Liquid crystal display |
US5717422A (en) * | 1994-01-25 | 1998-02-10 | Fergason; James L. | Variable intensity high contrast passive display |
US5592193A (en) * | 1994-03-10 | 1997-01-07 | Chunghwa Picture Tubes, Ltd. | Backlighting arrangement for LCD display panel |
JP3187669B2 (en) | 1994-04-01 | 2001-07-11 | 日本碍子株式会社 | Display element and display device |
JP3027298B2 (en) * | 1994-05-31 | 2000-03-27 | シャープ株式会社 | Liquid crystal display with backlight control function |
ATE349024T1 (en) * | 1994-08-04 | 2007-01-15 | Texas Instruments Inc | DISPLAY DEVICE |
US5639158A (en) * | 1994-08-19 | 1997-06-17 | Nec Corporation | Led-array light source |
US6184969B1 (en) | 1994-10-25 | 2001-02-06 | James L. Fergason | Optical display system and method, active and passive dithering using birefringence, color image superpositioning and display enhancement |
US5537256A (en) * | 1994-10-25 | 1996-07-16 | Fergason; James L. | Electronic dithering system using birefrigence for optical displays and method |
US6243055B1 (en) * | 1994-10-25 | 2001-06-05 | James L. Fergason | Optical display system and method with optical shifting of pixel position including conversion of pixel layout to form delta to stripe pattern by time base multiplexing |
US5572341A (en) * | 1994-10-25 | 1996-11-05 | Fergason; James L. | Electro-optical dithering system using birefringence for optical displays and method |
US5715029A (en) * | 1994-10-25 | 1998-02-03 | Fergason; James L. | Optical dithering system using birefringence for optical displays and method |
US6560018B1 (en) * | 1994-10-27 | 2003-05-06 | Massachusetts Institute Of Technology | Illumination system for transmissive light valve displays |
US5646702A (en) * | 1994-10-31 | 1997-07-08 | Honeywell Inc. | Field emitter liquid crystal display |
WO1996014206A1 (en) * | 1994-11-08 | 1996-05-17 | Spectra Science Corporation | Semiconductor nanocrystal display materials and display apparatus employing same |
JP3065494B2 (en) * | 1994-12-02 | 2000-07-17 | 東芝ライテック株式会社 | Fluorescent lamp and color liquid crystal display using the same |
US5658829A (en) * | 1995-02-21 | 1997-08-19 | Micron Technology, Inc. | Semiconductor processing method of forming an electrically conductive contact plug |
JP3764504B2 (en) * | 1995-02-28 | 2006-04-12 | ソニー株式会社 | Liquid crystal display |
US6111560A (en) * | 1995-04-18 | 2000-08-29 | Cambridge Display Technology Limited | Display with a light modulator and a light source |
JPH08334742A (en) | 1995-06-07 | 1996-12-17 | Canon Inc | Display device |
US5787030A (en) * | 1995-07-05 | 1998-07-28 | Sun Microsystems, Inc. | Correct and efficient sticky bit calculation for exact floating point divide/square root results |
US6120839A (en) * | 1995-07-20 | 2000-09-19 | E Ink Corporation | Electro-osmotic displays and materials for making the same |
US6120588A (en) * | 1996-07-19 | 2000-09-19 | E Ink Corporation | Electronically addressable microencapsulated ink and display thereof |
US5666174A (en) * | 1995-08-11 | 1997-09-09 | Cupolo, Iii; Anthony M. | Emissive liquid crystal display with liquid crystal between radiation source and phosphor layer |
US5737045A (en) | 1995-09-22 | 1998-04-07 | Ois Optical Imaging Systems, Inc. | LCD with notch filter |
US5754159A (en) | 1995-11-20 | 1998-05-19 | Texas Instruments Incorporated | Integrated liquid crystal display and backlight system for an electronic apparatus |
US5809215A (en) * | 1996-04-18 | 1998-09-15 | Lexmark International, Inc. | Method of printing to inhibit intercolor bleeding |
US5729242A (en) * | 1996-05-08 | 1998-03-17 | Hughes Electronics | Dual PDLC-projection head-up display |
US6323989B1 (en) * | 1996-07-19 | 2001-11-27 | E Ink Corporation | Electrophoretic displays using nanoparticles |
GB2317290B (en) * | 1996-09-11 | 2000-12-06 | Seos Displays Ltd | Image display apparatus |
KR100286828B1 (en) | 1996-09-18 | 2001-04-16 | 니시무로 타이죠 | Flat panel display device |
KR100261214B1 (en) * | 1997-02-27 | 2000-07-01 | 윤종용 | Histrogram equalization method and apparatus of a contrast expanding apparatus in image processing system |
JPH10269802A (en) | 1997-03-24 | 1998-10-09 | Sony Corp | Lighting system and image display unit |
US5986628A (en) * | 1997-05-14 | 1999-11-16 | Planar Systems, Inc. | Field sequential color AMEL display |
US5959777A (en) * | 1997-06-10 | 1999-09-28 | The University Of British Columbia | Passive high efficiency variable reflectivity image display device |
US6215920B1 (en) * | 1997-06-10 | 2001-04-10 | The University Of British Columbia | Electrophoretic, high index and phase transition control of total internal reflection in high efficiency variable reflectivity image displays |
JP3787983B2 (en) | 1997-06-18 | 2006-06-21 | セイコーエプソン株式会社 | Optical switching element, image display device, and projection device |
JPH1152412A (en) | 1997-07-31 | 1999-02-26 | Sony Corp | Reflection type liquid crystal display element |
JPH1164820A (en) | 1997-08-20 | 1999-03-05 | Nec Corp | Flat display device |
US6130774A (en) * | 1998-04-27 | 2000-10-10 | E Ink Corporation | Shutter mode microencapsulated electrophoretic display |
US6300932B1 (en) * | 1997-08-28 | 2001-10-09 | E Ink Corporation | Electrophoretic displays with luminescent particles and materials for making the same |
GB2330471A (en) | 1997-10-15 | 1999-04-21 | Secr Defence | Production of moving images for holography |
US6476783B2 (en) * | 1998-02-17 | 2002-11-05 | Sarnoff Corporation | Contrast enhancement for an electronic display device by using a black matrix and lens array on outer surface of display |
CA2328235A1 (en) | 1998-04-14 | 1999-10-21 | Halliburton Energy Services, Inc. | Methods and compositions for delaying the crosslinking of crosslinkable polysaccharide-based lost circulation materials |
JP3280307B2 (en) * | 1998-05-11 | 2002-05-13 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Liquid crystal display |
US6243068B1 (en) * | 1998-05-29 | 2001-06-05 | Silicon Graphics, Inc. | Liquid crystal flat panel display with enhanced backlight brightness and specially selected light sources |
US6864626B1 (en) | 1998-06-03 | 2005-03-08 | The Regents Of The University Of California | Electronic displays using optically pumped luminescent semiconductor nanocrystals |
US20050146258A1 (en) | 1999-06-02 | 2005-07-07 | Shimon Weiss | Electronic displays using optically pumped luminescent semiconductor nanocrystals |
JP3763378B2 (en) | 1998-07-21 | 2006-04-05 | シャープ株式会社 | Light guide film manufacturing method, light guide film manufactured by the manufacturing method, laminated film, and liquid crystal display device |
WO2000005703A1 (en) * | 1998-07-24 | 2000-02-03 | Seiko Epson Corporation | Display |
US6608439B1 (en) * | 1998-09-22 | 2003-08-19 | Emagin Corporation | Inorganic-based color conversion matrix element for organic color display devices and method of fabrication |
US6282313B1 (en) * | 1998-09-28 | 2001-08-28 | Eastman Kodak Company | Using a set of residual images to represent an extended color gamut digital image |
US6335983B1 (en) * | 1998-09-28 | 2002-01-01 | Eastman Kodak Company | Representing an extended color gamut digital image in a limited color gamut color space |
US6282312B1 (en) * | 1998-09-28 | 2001-08-28 | Eastman Kodak Company | System using one or more residual image(s) to represent an extended color gamut digital image |
US6282311B1 (en) * | 1998-09-28 | 2001-08-28 | Eastman Kodak Company | Using a residual image to represent an extended color gamut digital image |
US6285784B1 (en) * | 1998-09-28 | 2001-09-04 | Eastman Kodak Company | Method of applying manipulations to an extended color gamut digital image |
US6559826B1 (en) * | 1998-11-06 | 2003-05-06 | Silicon Graphics, Inc. | Method for modeling and updating a colorimetric reference profile for a flat panel display |
GB9828287D0 (en) | 1998-12-23 | 1999-02-17 | Secr Defence Brit | Image display system |
US6381372B1 (en) * | 1998-12-30 | 2002-04-30 | Xerox Corporation | Systems and methods for designing image processing filters using templates |
US6831624B1 (en) | 1999-01-15 | 2004-12-14 | Sharp Kabushiki Kaisha | Time sequentially scanned display |
JP2000214827A (en) | 1999-01-21 | 2000-08-04 | Toray Ind Inc | Color liquid crystal display device in field sequential drive system |
US6520646B2 (en) * | 1999-03-03 | 2003-02-18 | 3M Innovative Properties Company | Integrated front projection system with distortion correction and associated method |
JP2000275595A (en) | 1999-03-25 | 2000-10-06 | Sharp Corp | Method for inspection of liquid crystal display device |
US6439731B1 (en) * | 1999-04-05 | 2002-08-27 | Honeywell International, Inc. | Flat panel liquid crystal display |
US6327072B1 (en) * | 1999-04-06 | 2001-12-04 | E Ink Corporation | Microcell electrophoretic displays |
US6483643B1 (en) * | 1999-04-08 | 2002-11-19 | Larry Zuchowski | Controlled gain projection screen |
US6600467B1 (en) * | 1999-04-28 | 2003-07-29 | Homer L. Webb | Flat panel display architecture |
US6144162A (en) * | 1999-04-28 | 2000-11-07 | Intel Corporation | Controlling polymer displays |
US7071907B1 (en) | 1999-05-07 | 2006-07-04 | Candescent Technologies Corporation | Display with active contrast enhancement |
US6795585B1 (en) | 1999-07-16 | 2004-09-21 | Eastman Kodak Company | Representing digital images in a plurality of image processing states |
US6631995B2 (en) | 1999-09-02 | 2003-10-14 | Koninklijke Philips Electronics N.V. | Method of and device for generating an image having a desired brightness |
JP2001100699A (en) | 1999-09-29 | 2001-04-13 | Canon Inc | Projection display device and its application system |
JP2001100689A (en) | 1999-09-30 | 2001-04-13 | Canon Inc | Display device |
JP3688574B2 (en) | 1999-10-08 | 2005-08-31 | シャープ株式会社 | Liquid crystal display device and light source device |
US6054120A (en) * | 1999-10-08 | 2000-04-25 | Burgoyne; Bradley C. | Sunscreen applicator system |
JP4519251B2 (en) * | 1999-10-13 | 2010-08-04 | シャープ株式会社 | Liquid crystal display device and control method thereof |
JP2001265296A (en) | 2000-01-14 | 2001-09-28 | Sharp Corp | Transmission type liquid crystal display device and picture processing method |
US6301393B1 (en) * | 2000-01-21 | 2001-10-09 | Eastman Kodak Company | Using a residual image formed from a clipped limited color gamut digital image to represent an extended color gamut digital image |
US6414661B1 (en) * | 2000-02-22 | 2002-07-02 | Sarnoff Corporation | Method and apparatus for calibrating display devices and automatically compensating for loss in their efficiency over time |
EP1202244A4 (en) | 2000-03-14 | 2005-08-31 | Mitsubishi Electric Corp | Image display and image displaying method |
US7224335B2 (en) | 2000-03-15 | 2007-05-29 | Imax Corporation | DMD-based image display systems |
EP1136874A1 (en) | 2000-03-22 | 2001-09-26 | Hewlett-Packard Company, A Delaware Corporation | Projection screen |
US6748106B1 (en) | 2000-03-28 | 2004-06-08 | Eastman Kodak Company | Method for representing an extended color gamut digital image on a hard-copy output medium |
US6428189B1 (en) * | 2000-03-31 | 2002-08-06 | Relume Corporation | L.E.D. thermal management |
US6822760B1 (en) | 2000-04-05 | 2004-11-23 | Eastman Kodak Company | Method of processing and paying for an extended color gamut digital image |
TWI240241B (en) * | 2000-05-04 | 2005-09-21 | Koninkl Philips Electronics Nv | Assembly of a display device and an illumination system |
US6621482B2 (en) * | 2000-05-15 | 2003-09-16 | Koninklijke Philips Electronics N.V. | Display arrangement with backlight means |
US6608614B1 (en) * | 2000-06-22 | 2003-08-19 | Rockwell Collins, Inc. | Led-based LCD backlight with extended color space |
ATE538594T1 (en) | 2000-07-03 | 2012-01-15 | Imax Corp | METHOD AND DEVICE FOR EXPANDING THE DYNAMIC RANGE OF A PROJECTION SYSTEM |
US6775407B1 (en) | 2000-08-02 | 2004-08-10 | Eastman Kodak Company | Producing a final modified digital image using a source digital image and a difference digital image |
US6754384B1 (en) | 2000-08-30 | 2004-06-22 | Eastman Kodak Company | Method for processing an extended color gamut digital image using an image information parameter |
US6952195B2 (en) | 2000-09-12 | 2005-10-04 | Fuji Photo Film Co., Ltd. | Image display device |
JP2002091385A (en) | 2000-09-12 | 2002-03-27 | Matsushita Electric Ind Co Ltd | Illuminator |
JP3523170B2 (en) | 2000-09-21 | 2004-04-26 | 株式会社東芝 | Display device |
US6680834B2 (en) | 2000-10-04 | 2004-01-20 | Honeywell International Inc. | Apparatus and method for controlling LED arrays |
JP2002140338A (en) | 2000-10-31 | 2002-05-17 | Toshiba Corp | Device and method for supporting construction of dictionary |
US6644832B2 (en) | 2000-12-25 | 2003-11-11 | Seiko Epson Corporation | Illumination device and manufacturing method therefor, display device, and electronic instrument |
US6930737B2 (en) | 2001-01-16 | 2005-08-16 | Visteon Global Technologies, Inc. | LED backlighting system |
TW548964B (en) | 2001-01-24 | 2003-08-21 | Koninkl Philips Electronics Nv | Window brightness enhancement for LCD display |
US20020110180A1 (en) * | 2001-02-09 | 2002-08-15 | Barney Alfred A. | Temperature-sensing composition |
EP2267520B1 (en) | 2001-02-27 | 2018-07-25 | Dolby Laboratories Licensing Corporation | A method and device for displaying an image |
US20020159002A1 (en) * | 2001-03-30 | 2002-10-31 | Koninklijke Philips Electronics N.V. | Direct backlighting for liquid crystal displays |
US6844903B2 (en) * | 2001-04-04 | 2005-01-18 | Lumileds Lighting U.S., Llc | Blue backlight and phosphor layer for a color LCD |
US6590561B1 (en) * | 2001-05-26 | 2003-07-08 | Garmin Ltd. | Computer program, method, and device for controlling the brightness of a display |
US6863401B2 (en) | 2001-06-30 | 2005-03-08 | Texas Instruments Incorporated | Illumination system |
JP2003027057A (en) * | 2001-07-17 | 2003-01-29 | Hitachi Ltd | Light source and image display device using the same |
DE10137042A1 (en) * | 2001-07-31 | 2003-02-20 | Patent Treuhand Ges Fuer Elektrische Gluehlampen Mbh | Planar light source based on LED |
US7002533B2 (en) * | 2001-08-17 | 2006-02-21 | Michel Sayag | Dual-stage high-contrast electronic image display |
GB2379317A (en) | 2001-08-30 | 2003-03-05 | Cambridge Display Tech Ltd | Optoelectronic display operating by photoluminescence quenching |
US7175281B1 (en) | 2003-05-13 | 2007-02-13 | Lightmaster Systems, Inc. | Method and apparatus to increase the contrast ratio of the image produced by a LCoS based light engine |
CN101241684A (en) | 2001-11-02 | 2008-08-13 | 夏普株式会社 | Image display device |
US7064740B2 (en) * | 2001-11-09 | 2006-06-20 | Sharp Laboratories Of America, Inc. | Backlit display with improved dynamic range |
US7015991B2 (en) | 2001-12-21 | 2006-03-21 | 3M Innovative Properties Company | Color pre-filter for single-panel projection display system |
WO2003058726A1 (en) | 2001-12-28 | 2003-07-17 | Sanken Electric Co., Ltd. | Semiconductor light-emitting device, light-emitting display, method for manufacturing semiconductor light-emitting device, and method for manufacturing light-emitting display |
ATE448549T1 (en) * | 2002-01-11 | 2009-11-15 | Texas Instruments Inc | SPATIAL LIGHT MODULATOR WITH CHARGE PUMP PIXEL CELL |
US6720942B2 (en) * | 2002-02-12 | 2004-04-13 | Eastman Kodak Company | Flat-panel light emitting pixel with luminance feedback |
ES2675880T3 (en) | 2002-03-13 | 2018-07-13 | Dolby Laboratories Licensing Corporation | Failure compensation of light emitting element on a monitor |
US6802612B2 (en) | 2002-03-15 | 2004-10-12 | Hewlett-Packard Development Company, L.P. | Configurations for color displays by the use of lenticular optics |
JP2003346530A (en) | 2002-05-23 | 2003-12-05 | Nippon Sheet Glass Co Ltd | Planar light source and image scanner |
US6728023B1 (en) | 2002-05-28 | 2004-04-27 | Silicon Light Machines | Optical device arrays with optimized image resolution |
US6753661B2 (en) | 2002-06-17 | 2004-06-22 | Koninklijke Philips Electronics N.V. | LED-based white-light backlighting for electronic displays |
NZ517713A (en) | 2002-06-25 | 2005-03-24 | Puredepth Ltd | Enhanced viewing experience of a display through localised dynamic control of background lighting level |
US20040012551A1 (en) | 2002-07-16 | 2004-01-22 | Takatoshi Ishii | Adaptive overdrive and backlight control for TFT LCD pixel accelerator |
AU2003247014A1 (en) | 2002-07-23 | 2004-02-09 | Koninklijke Philips Electronics N.V. | Electroluminescent display, electronic device comprising such a display and method of manufacturing an electroluminescent display |
KR100828531B1 (en) | 2002-07-26 | 2008-05-13 | 삼성전자주식회사 | Liquid crystal display |
US6832037B2 (en) | 2002-08-09 | 2004-12-14 | Eastman Kodak Company | Waveguide and method of making same |
US7036946B1 (en) * | 2002-09-13 | 2006-05-02 | Rockwell Collins, Inc. | LCD backlight with UV light-emitting diodes and planar reactive element |
US6817717B2 (en) | 2002-09-19 | 2004-11-16 | Hewlett-Packard Development Company, L.P. | Display system with low and high resolution modulators |
DE10245892A1 (en) | 2002-09-30 | 2004-05-13 | Siemens Ag | Illumination device for backlighting an image display device |
KR100712334B1 (en) | 2002-09-30 | 2007-05-02 | 엘지전자 주식회사 | Method for controling a brightness level of LCD |
US7430022B2 (en) | 2002-10-01 | 2008-09-30 | Koninklijke Philips Electronics N.V. | Color display device |
JP4087681B2 (en) | 2002-10-29 | 2008-05-21 | 株式会社日立製作所 | LIGHTING DEVICE AND DISPLAY DEVICE USING THE SAME |
GB0228089D0 (en) | 2002-12-02 | 2003-01-08 | Seos Ltd | Dynamic range enhancement of image display apparatus |
JP2004184852A (en) * | 2002-12-05 | 2004-07-02 | Olympus Corp | Display device, light source device and illuminator |
JP3498290B1 (en) | 2002-12-19 | 2004-02-16 | 俊二 岸村 | White LED lighting device |
EP1579733B1 (en) | 2002-12-26 | 2008-04-09 | Koninklijke Philips Electronics N.V. | Color temperature correction for phosphor converted leds |
KR100852579B1 (en) | 2003-03-31 | 2008-08-14 | 샤프 가부시키가이샤 | Surface illumination device and liquid display device using the same |
JP2004325647A (en) | 2003-04-23 | 2004-11-18 | Sharp Corp | Display element |
AU2004235139A1 (en) | 2003-04-25 | 2004-11-11 | Visioneered Image Systems, Inc. | Led illumination source/display with individual led brightness monitoring capability and calibration method |
US7289163B2 (en) | 2003-04-28 | 2007-10-30 | Samsung Electronics Co., Ltd. | Method and apparatus for adjusting color edge center in color transient improvement |
EP1640787B1 (en) | 2003-06-20 | 2009-04-01 | Sharp Kabushiki Kaisha | Display |
US7097495B2 (en) | 2003-07-14 | 2006-08-29 | Tribotek, Inc. | System and methods for connecting electrical components |
EP1648038B1 (en) | 2003-07-22 | 2011-02-16 | NGK Insulators, Ltd. | Actuator element and device having actuator element |
US7052152B2 (en) | 2003-10-03 | 2006-05-30 | Philips Lumileds Lighting Company, Llc | LCD backlight using two-dimensional array LEDs |
US20070024576A1 (en) | 2004-01-13 | 2007-02-01 | Hassan Paddy A | Correction arrangements for portable devices with oled displays |
GB2410116A (en) * | 2004-01-17 | 2005-07-20 | Sharp Kk | Illumination system and display device |
JP4628770B2 (en) * | 2004-02-09 | 2011-02-09 | 株式会社日立製作所 | Image display device having illumination device and image display method |
JP4139344B2 (en) | 2004-03-15 | 2008-08-27 | シャープ株式会社 | Display device |
US7354172B2 (en) | 2004-03-15 | 2008-04-08 | Philips Solid-State Lighting Solutions, Inc. | Methods and apparatus for controlled lighting based on a reference gamut |
EP1745436B1 (en) | 2004-04-15 | 2012-05-30 | Dolby Laboratories Licensing Corporation | Methods and systems for converting images from low dynamic range to high dynamic range |
US7532192B2 (en) | 2004-05-04 | 2009-05-12 | Sharp Laboratories Of America, Inc. | Liquid crystal display with filtered black point |
US7768023B2 (en) | 2005-10-14 | 2010-08-03 | The Regents Of The University Of California | Photonic structures for efficient light extraction and conversion in multi-color light emitting devices |
US7480042B1 (en) | 2004-06-30 | 2009-01-20 | Applied Biosystems Inc. | Luminescence reference standards |
KR20070039539A (en) * | 2004-07-15 | 2007-04-12 | 소니 가부시끼 가이샤 | Color filter and color liquid crystal display device |
US8217970B2 (en) * | 2004-07-27 | 2012-07-10 | Dolby Laboratories Licensing Corporation | Rapid image rendering on dual-modulator displays |
CN100507988C (en) | 2004-07-27 | 2009-07-01 | 杜比实验室特许公司 | Rapid image rendering on dual-modulator displays |
US7575697B2 (en) * | 2004-08-04 | 2009-08-18 | Intematix Corporation | Silicate-based green phosphors |
US7113670B2 (en) | 2004-09-15 | 2006-09-26 | Research In Motion Limited | Method and device to improve backlight uniformity |
JP2006114909A (en) | 2004-10-14 | 2006-04-27 | Agilent Technol Inc | Flash module |
US20060092183A1 (en) | 2004-10-22 | 2006-05-04 | Amedeo Corporation | System and method for setting brightness uniformity in an active-matrix organic light-emitting diode (OLED) flat-panel display |
US7481562B2 (en) | 2004-11-18 | 2009-01-27 | Avago Technologies Ecbu Ip (Singapore) Pte. Ltd. | Device and method for providing illuminating light using quantum dots |
KR100735148B1 (en) | 2004-11-22 | 2007-07-03 | (주)케이디티 | Backlight unit by phosphorescent diffusion sheet |
TWI263802B (en) | 2004-12-03 | 2006-10-11 | Innolux Display Corp | Color filter |
JP5084111B2 (en) | 2005-03-31 | 2012-11-28 | 三洋電機株式会社 | Display device and driving method of display device |
US7791561B2 (en) | 2005-04-01 | 2010-09-07 | Prysm, Inc. | Display systems having screens with optical fluorescent materials |
US20060221022A1 (en) | 2005-04-01 | 2006-10-05 | Roger Hajjar | Laser vector scanner systems with display screens having optical fluorescent materials |
CN101218621B (en) | 2005-04-01 | 2011-07-13 | Prysm公司 | Display systems and devices having screens with optical fluorescent materials |
JP4432818B2 (en) | 2005-04-01 | 2010-03-17 | セイコーエプソン株式会社 | Image display device, image display method, and image display program |
US7334901B2 (en) | 2005-04-22 | 2008-02-26 | Ostendo Technologies, Inc. | Low profile, large screen display using a rear projection array system |
JP2006309219A (en) | 2005-04-25 | 2006-11-09 | Samsung Electronics Co Ltd | Photo-luminescence liquid crystal display |
US8000005B2 (en) | 2006-03-31 | 2011-08-16 | Prysm, Inc. | Multilayered fluorescent screens for scanning beam display systems |
JP5057692B2 (en) * | 2005-04-27 | 2012-10-24 | サムソン エルイーディー カンパニーリミテッド. | Backlight unit using light emitting diode |
JP2006309238A (en) | 2005-04-27 | 2006-11-09 | Samsung Electronics Co Ltd | Photoluminescence liquid crystal display |
KR101110071B1 (en) * | 2005-04-29 | 2012-02-24 | 삼성전자주식회사 | Photo-Luminescenct Liquid Crystal Display |
KR101110072B1 (en) * | 2005-06-02 | 2012-02-24 | 삼성전자주식회사 | Photo-Luminescenct Liquid Crystal Display |
US8718437B2 (en) | 2006-03-07 | 2014-05-06 | Qd Vision, Inc. | Compositions, optical component, system including an optical component, devices, and other products |
US8215815B2 (en) | 2005-06-07 | 2012-07-10 | Oree, Inc. | Illumination apparatus and methods of forming the same |
US7404645B2 (en) | 2005-06-20 | 2008-07-29 | Digital Display Innovations, Llc | Image and light source modulation for a digital display system |
US7733017B2 (en) | 2005-07-08 | 2010-06-08 | Peysakh Shapiro | Display apparatus with replaceable electroluminescent element |
US7513669B2 (en) | 2005-08-01 | 2009-04-07 | Avago Technologies General Ip (Singapore) Pte. Ltd. | Light source for LCD back-lit displays |
TWI271883B (en) | 2005-08-04 | 2007-01-21 | Jung-Chieh Su | Light-emitting devices with high extraction efficiency |
ATE514198T1 (en) | 2005-08-15 | 2011-07-15 | Koninkl Philips Electronics Nv | LIGHT SOURCE AND METHOD FOR GENERATING LIGHT WITH INDEPENDENTLY CHANGING COLOR AND BRIGHTNESS |
CN100517016C (en) | 2005-10-27 | 2009-07-22 | 鸿富锦精密工业(深圳)有限公司 | Light source and backlight module |
US7321193B2 (en) | 2005-10-31 | 2008-01-22 | Osram Opto Semiconductors Gmbh | Device structure for OLED light device having multi element light extraction and luminescence conversion layer |
US7420323B2 (en) | 2005-10-31 | 2008-09-02 | Osram Opto Semiconductors Gmbh | Electroluminescent apparatus having a structured luminescence conversion layer |
US7486304B2 (en) | 2005-12-21 | 2009-02-03 | Nokia Corporation | Display device with dynamic color gamut |
TWI273285B (en) | 2005-12-23 | 2007-02-11 | Wintek Corp | Color filter having capability of changing light-color |
US7486854B2 (en) | 2006-01-24 | 2009-02-03 | Uni-Pixel Displays, Inc. | Optical microstructures for light extraction and control |
US7486354B2 (en) | 2006-01-26 | 2009-02-03 | Hannstar Display Corp. | Backlight module of a liquid crystal display, display device, method of improving color gamut of a display device |
CA2641310C (en) | 2006-02-03 | 2013-08-20 | Imclone Systems Incorporated | Igf-ir antagonists as adjuvants for treatment of prostate cancer |
WO2007114918A2 (en) | 2006-04-04 | 2007-10-11 | Microvision, Inc. | Electronic display with photoluminescent wavelength conversion |
KR100783251B1 (en) | 2006-04-10 | 2007-12-06 | 삼성전기주식회사 | Multi-Layered White Light Emitting Diode Using Quantum Dots and Method of Preparing The Same |
US20070247573A1 (en) | 2006-04-19 | 2007-10-25 | 3M Innovative Properties Company | Transflective LC Display Having Narrow Band Backlight and Spectrally Notched Transflector |
KR100790698B1 (en) | 2006-04-19 | 2008-01-02 | 삼성전기주식회사 | Backlight unit for liquid crystal display device |
US20070268240A1 (en) | 2006-05-19 | 2007-11-22 | Lee Sang-Jin | Display device and method of driving the display device |
KR100759398B1 (en) * | 2006-06-20 | 2007-09-19 | 삼성에스디아이 주식회사 | Light emission device and liquid crystal display device using the same as back light unit |
US7880381B2 (en) | 2006-07-05 | 2011-02-01 | Avago Technologies General Ip (Singapore) Pte. Ltd. | LED with light absorbing encapsulant and related methodology |
US8947619B2 (en) * | 2006-07-06 | 2015-02-03 | Intematix Corporation | Photoluminescence color display comprising quantum dots material and a wavelength selective filter that allows passage of excitation radiation and prevents passage of light generated by photoluminescence materials |
US20080074583A1 (en) * | 2006-07-06 | 2008-03-27 | Intematix Corporation | Photo-luminescence color liquid crystal display |
KR101204861B1 (en) | 2006-07-28 | 2012-11-26 | 삼성디스플레이 주식회사 | Backlight unit and liquid crystal display comprising the same |
KR100828366B1 (en) | 2006-08-01 | 2008-05-08 | 삼성전자주식회사 | LCD TV having dimming panel and driving method therefor |
WO2008021962A2 (en) * | 2006-08-11 | 2008-02-21 | Massachusetts Institute Of Technology | Blue light emitting semiconductor nanocrystals and devices |
US7703942B2 (en) * | 2006-08-31 | 2010-04-27 | Rensselaer Polytechnic Institute | High-efficient light engines using light emitting diodes |
US7751663B2 (en) | 2006-09-21 | 2010-07-06 | Uni-Pixel Displays, Inc. | Backside reflection optical display |
CN101563791B (en) | 2006-09-27 | 2011-09-07 | 株式会社东芝 | Semiconductor light emitting device, backlight composed of the semiconductor light emitting device, and display device |
GB2442505A (en) | 2006-10-04 | 2008-04-09 | Sharp Kk | A display with a primary light source for illuminating a nanophosphor re-emission material |
JP4851908B2 (en) | 2006-10-10 | 2012-01-11 | 株式会社 日立ディスプレイズ | Liquid crystal display |
KR101361861B1 (en) | 2006-11-08 | 2014-02-12 | 엘지디스플레이 주식회사 | Organic light emitting diodes and method of manufacturing the same |
WO2008065575A1 (en) | 2006-11-30 | 2008-06-05 | Nxp B.V. | Device and method for processing color image data |
JP2008145551A (en) | 2006-12-06 | 2008-06-26 | Sony Corp | Display device |
EP2092796A4 (en) | 2006-12-11 | 2016-11-16 | Philips Lighting Holding Bv | Luminaire control system and method |
KR20080058820A (en) | 2006-12-22 | 2008-06-26 | 삼성전자주식회사 | Display apparatus and control method thereof |
KR20080058821A (en) | 2006-12-22 | 2008-06-26 | 삼성전자주식회사 | Backlight unit and liquid crystal display |
US7845822B2 (en) | 2006-12-29 | 2010-12-07 | Koninklijke Philips Electronics N.V. | Illumination device including a color selecting panel for recycling unwanted light |
KR100946015B1 (en) * | 2007-01-02 | 2010-03-09 | 삼성전기주식회사 | White led device and light source module for lcd backlight using the same |
US20080170176A1 (en) * | 2007-01-11 | 2008-07-17 | Vastview Technology Inc. | Backlight Module Having Phosphor Layer and Liquid Crystal Display Device Using the Same |
US20080172197A1 (en) | 2007-01-11 | 2008-07-17 | Motorola, Inc. | Single laser multi-color projection display with quantum dot screen |
WO2008094153A1 (en) * | 2007-01-31 | 2008-08-07 | Dolby Laboratories Licensing Corporation | Multiple modulator displays and related methods |
DE102007009530A1 (en) | 2007-02-27 | 2008-08-28 | Osram Opto Semiconductors Gmbh | Organic light-emitting diode for lighting purposes predominantly emitting white light mixed with colors and composite video signal conversation, comprises substrate layer structure, anode, cathode and intermediate arranged functional layer |
CN101627482A (en) | 2007-03-08 | 2010-01-13 | 3M创新有限公司 | Array of luminescent elements |
US7478922B2 (en) | 2007-03-14 | 2009-01-20 | Renaissance Lighting, Inc. | Set-point validation for color/intensity settings of light fixtures |
US20100155749A1 (en) | 2007-03-19 | 2010-06-24 | Nanosys, Inc. | Light-emitting diode (led) devices comprising nanocrystals |
US7687816B2 (en) | 2007-03-20 | 2010-03-30 | International Business Machines Corporation | Light emitting diode |
US9279079B2 (en) * | 2007-05-30 | 2016-03-08 | Sharp Kabushiki Kaisha | Method of manufacturing phosphor, light-emitting device, and image display apparatus |
CN201062757Y (en) | 2007-06-05 | 2008-05-21 | 诸建平 | Illuminating device of white light surface light source |
TWM322627U (en) * | 2007-06-06 | 2007-11-21 | Acpa Energy Conversion Devices | Passive light-emitting module whose visible lights are excited from the ultraviolet |
KR101730164B1 (en) | 2007-07-18 | 2017-04-25 | 삼성전자주식회사 | Quantum dot-based light sheets useful for solid-state lighting |
WO2009014707A2 (en) | 2007-07-23 | 2009-01-29 | Qd Vision, Inc. | Quantum dot light enhancement substrate and lighting device including same |
US8585273B2 (en) | 2007-07-31 | 2013-11-19 | Rambus Delaware Llc | Illumination assembly including wavelength converting material |
TWI345671B (en) | 2007-08-10 | 2011-07-21 | Au Optronics Corp | Thin film transistor, pixel structure and liquid crystal display panel |
US8128249B2 (en) | 2007-08-28 | 2012-03-06 | Qd Vision, Inc. | Apparatus for selectively backlighting a material |
US7934862B2 (en) * | 2007-09-24 | 2011-05-03 | Munisamy Anandan | UV based color pixel backlight for liquid crystal display |
CN102648435A (en) | 2007-09-27 | 2012-08-22 | 夏普株式会社 | Display device |
WO2009041594A1 (en) | 2007-09-28 | 2009-04-02 | Dai Nippon Printing Co., Ltd. | Electroluminescence element |
KR101376755B1 (en) | 2007-10-09 | 2014-03-24 | 삼성디스플레이 주식회사 | Display Device |
KR101415566B1 (en) | 2007-10-29 | 2014-07-04 | 삼성디스플레이 주식회사 | Display device |
JP4613947B2 (en) | 2007-12-07 | 2011-01-19 | ソニー株式会社 | Illumination device, color conversion element, and display device |
JP2009283438A (en) | 2007-12-07 | 2009-12-03 | Sony Corp | Lighting device, display device, and manufacturing method of lighting device |
JP5134618B2 (en) | 2007-12-18 | 2013-01-30 | Idec株式会社 | Wavelength converter and light emitting device |
KR101460155B1 (en) | 2008-01-15 | 2014-11-10 | 삼성전자주식회사 | Backlight unit and liquid crystal display having the same |
US8029139B2 (en) | 2008-01-29 | 2011-10-04 | Eastman Kodak Company | 2D/3D switchable color display apparatus with narrow band emitters |
US20090194774A1 (en) | 2008-02-04 | 2009-08-06 | Kismart Corporation | Light source module with wavelength converting structure and the method of forming the same |
US7832885B2 (en) | 2008-02-05 | 2010-11-16 | Kismart Corporation | Patterned wavelength converting structure |
BRPI0822306A2 (en) | 2008-02-14 | 2015-06-16 | Sharp Kk | Display device |
TW200938913A (en) | 2008-03-13 | 2009-09-16 | Kismart Corp | A flat panel display capable of multi-sided viewings and its back light module |
JP2009251129A (en) | 2008-04-02 | 2009-10-29 | Optoelectronic Industry & Technology Development Association | Color filter for liquid crystal display device and liquid crystal display device |
JP5369486B2 (en) | 2008-04-28 | 2013-12-18 | 豊田合成株式会社 | Light emitting device |
EP2120448A1 (en) | 2008-05-14 | 2009-11-18 | Thomson Licensing | Method of processing of a compressed image into a gamut mapped image using spatial frequency analysis |
US8246408B2 (en) | 2008-06-13 | 2012-08-21 | Barco, Inc. | Color calibration system for a video display |
US20090322800A1 (en) | 2008-06-25 | 2009-12-31 | Dolby Laboratories Licensing Corporation | Method and apparatus in various embodiments for hdr implementation in display devices |
US7988311B2 (en) | 2008-06-30 | 2011-08-02 | Bridgelux, Inc. | Light emitting device having a phosphor layer |
US8459855B2 (en) * | 2008-07-28 | 2013-06-11 | Munisamy Anandan | UV LED based color pixel backlight incorporating quantum dots for increasing color gamut of LCD |
TW201007321A (en) * | 2008-08-08 | 2010-02-16 | Wintek Corp | Electro-wetting display device |
US8314767B2 (en) | 2008-08-30 | 2012-11-20 | Sharp Laboratories Of America, Inc. | Methods and systems for reducing view-angle-induced color shift |
EP2164302A1 (en) | 2008-09-12 | 2010-03-17 | Ilford Imaging Switzerland Gmbh | Optical element and method for its production |
US7858409B2 (en) | 2008-09-18 | 2010-12-28 | Koninklijke Philips Electronics N.V. | White point compensated LEDs for LCD displays |
US8294848B2 (en) | 2008-10-01 | 2012-10-23 | Samsung Display Co., Ltd. | Liquid crystal display having light diffusion layer |
JP2010092705A (en) | 2008-10-08 | 2010-04-22 | Sony Corp | Illuminating device and display device using this |
KR101225574B1 (en) | 2008-10-14 | 2013-01-25 | 돌비 레버러토리즈 라이쎈싱 코오포레이션 | Backlight simulation at reduced resolutions to determine spatial modulation of light for high dynamic range images |
TWI416454B (en) | 2008-10-31 | 2013-11-21 | Dynascan Technology Corp | A method for compensating the uniformity of a liquid crystal display with a non - uniform backlight and the display |
US8363100B2 (en) | 2008-11-19 | 2013-01-29 | Honeywell International Inc. | Three dimensional display systems and methods for producing three dimensional images |
GB0821122D0 (en) | 2008-11-19 | 2008-12-24 | Nanoco Technologies Ltd | Semiconductor nanoparticle - based light emitting devices and associated materials and methods |
JP4772105B2 (en) * | 2008-12-10 | 2011-09-14 | シャープ株式会社 | Semiconductor light emitting device and image display device using the same |
KR101462658B1 (en) * | 2008-12-19 | 2014-11-17 | 삼성전자 주식회사 | Semiconductor Nanocrystal and Preparation Method thereof |
US8272770B2 (en) | 2009-01-02 | 2012-09-25 | Rambus International Ltd. | TIR switched flat panel display |
JP5367383B2 (en) | 2009-01-14 | 2013-12-11 | 株式会社東芝 | Display device and driving method thereof |
WO2010085505A1 (en) | 2009-01-21 | 2010-07-29 | Dolby Laboratories Licensing Corporation | Apparatus and methods for color displays |
KR101562022B1 (en) | 2009-02-02 | 2015-10-21 | 삼성디스플레이 주식회사 | Light emitting diode unit display device having the same and manufacturing mathod of the light emitting diode unit |
KR101584663B1 (en) | 2009-02-17 | 2016-01-13 | 삼성전자주식회사 | Polymer dispersed liquid crystal display apparatus using quantum dot |
KR101631986B1 (en) | 2009-02-18 | 2016-06-21 | 삼성전자주식회사 | Light guide plate and display apparatus employing the same |
US20100207865A1 (en) | 2009-02-19 | 2010-08-19 | Zoran Corporation | Systems and methods for display device backlight compensation |
US20100214282A1 (en) | 2009-02-24 | 2010-08-26 | Dolby Laboratories Licensing Corporation | Apparatus for providing light source modulation in dual modulator displays |
US9524700B2 (en) | 2009-05-14 | 2016-12-20 | Pure Depth Limited | Method and system for displaying images of various formats on a single display |
US8379039B2 (en) | 2009-06-07 | 2013-02-19 | Apple Inc. | Reformatting content with proper color-region conversion |
KR20110012246A (en) | 2009-07-30 | 2011-02-09 | 엘지이노텍 주식회사 | Backlight unit |
US9341887B2 (en) | 2009-09-11 | 2016-05-17 | Dolby Laboratories Licensing Corporation | Displays with a backlight incorporating reflecting layer |
KR20110041824A (en) | 2009-10-16 | 2011-04-22 | 엘지디스플레이 주식회사 | Display device using quantum dot |
KR101318444B1 (en) | 2009-11-23 | 2013-10-16 | 엘지디스플레이 주식회사 | Method of compensating pixel data and liquid crystal display |
KR101563478B1 (en) | 2009-12-22 | 2015-10-26 | 엘지이노텍 주식회사 | Backlight apparatus including quantum dots |
US20110205251A1 (en) | 2010-02-22 | 2011-08-25 | David Auld | Passive eyewear stereoscopic viewing system with frequency selective emitter |
TR201001777A2 (en) | 2010-03-09 | 2011-09-21 | Vestel Elektroni̇k Sanayi̇ Ve Ti̇caret Anoni̇m Şi̇rketi̇@ | Backlight unit and making method for liquid crystal display. |
US8294168B2 (en) | 2010-06-04 | 2012-10-23 | Samsung Electronics Co., Ltd. | Light source module using quantum dots, backlight unit employing the light source module, display apparatus, and illumination apparatus |
US8651684B2 (en) | 2010-07-28 | 2014-02-18 | Unipixel Displays, Inc. | Two and three-dimensional image with optical emission frequency control |
US8436549B2 (en) | 2010-08-13 | 2013-05-07 | Bridgelux, Inc. | Drive circuit for a color temperature tunable LED light source |
US20120050632A1 (en) | 2010-08-31 | 2012-03-01 | Chi Lin Technology Co., Ltd. | Display apparatus having quantum dot layer |
US8773477B2 (en) | 2010-09-15 | 2014-07-08 | Dolby Laboratories Licensing Corporation | Method and apparatus for edge lit displays |
US8736674B2 (en) | 2010-09-23 | 2014-05-27 | Dolby Laboratories Licensing Corporation | Method and system for 3D display calibration with feedback determined by a camera device |
US8994714B2 (en) | 2010-09-23 | 2015-03-31 | Dolby Laboratories Licensing Corporation | Method and system for display calibration with feedback determined by a camera device |
KR102381463B1 (en) | 2010-11-10 | 2022-04-01 | 나노시스, 인크. | Quantum dot films, lighting devices, and lighting methods |
US8514352B2 (en) | 2010-12-10 | 2013-08-20 | Sharp Kabushiki Kaisha | Phosphor-based display |
CN103262144B (en) | 2010-12-17 | 2016-08-17 | 杜比实验室特许公司 | Quantum dot for display floater |
KR20120078883A (en) | 2011-01-03 | 2012-07-11 | 엘지전자 주식회사 | Display apparatus |
KR101177480B1 (en) | 2011-02-14 | 2012-08-24 | 엘지전자 주식회사 | Lighting apparatus and display device comprising the same |
US9183811B2 (en) | 2011-04-01 | 2015-11-10 | Sharp Kabushiki Kaisha | Method of correcting unevenness of display panel and correction system |
KR20120131628A (en) | 2011-05-26 | 2012-12-05 | 삼성디스플레이 주식회사 | Display device |
KR101793741B1 (en) | 2011-06-23 | 2017-11-03 | 엘지이노텍 주식회사 | Display device |
US9082349B2 (en) | 2011-08-30 | 2015-07-14 | Sharp Laboratories Of America, Inc. | Multi-primary display with active backlight |
US8698980B2 (en) | 2011-11-14 | 2014-04-15 | Planck Co., Ltd. | Color regulating device for illumination and apparatus using the same, and method of regulating color |
JP2013161053A (en) | 2012-02-08 | 2013-08-19 | Nikon Corp | Image display device |
US20130215136A1 (en) | 2012-02-20 | 2013-08-22 | Apple Inc. | Liquid crystal display with large color gamut |
US20130335677A1 (en) | 2012-06-15 | 2013-12-19 | Apple Inc. | Quantum Dot-Enhanced Display Having Dichroic Filter |
US10680194B2 (en) * | 2015-01-12 | 2020-06-09 | Massachusetts Institute Of Technology | Transparent luminescent displays enabled by electric-field-induced quenching of photoluminescent pixels |
-
2010
- 2010-02-17 US US12/707,276 patent/US20100214282A1/en not_active Abandoned
- 2010-02-24 CA CA2694451A patent/CA2694451A1/en not_active Abandoned
-
2014
- 2014-03-17 US US14/215,856 patent/US9099046B2/en active Active
-
2015
- 2015-06-24 US US14/749,195 patent/US9478182B2/en active Active
-
2016
- 2016-10-19 US US15/298,094 patent/US9911389B2/en active Active
-
2018
- 2018-02-26 US US15/905,085 patent/US10373574B2/en active Active
Also Published As
Publication number | Publication date |
---|---|
US10373574B2 (en) | 2019-08-06 |
US20140198142A1 (en) | 2014-07-17 |
US9099046B2 (en) | 2015-08-04 |
US20100214282A1 (en) | 2010-08-26 |
US20180190215A1 (en) | 2018-07-05 |
US20150294630A1 (en) | 2015-10-15 |
US9478182B2 (en) | 2016-10-25 |
US20170039963A1 (en) | 2017-02-09 |
US9911389B2 (en) | 2018-03-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2694451A1 (en) | Cluster and discriminant analysis for vehicles detection | |
US8825586B2 (en) | Vehicle type recognition at a checkpoint using PCA and BIC | |
CN103235933B (en) | A kind of vehicle abnormality behavioral value method based on HMM | |
Yi et al. | A machine learning based personalized system for driving state recognition | |
CN101334845B (en) | Video frequency behaviors recognition method based on track sequence analysis and rule induction | |
CN104504400B (en) | A kind of driver's anomaly detection method based on online behavior modeling | |
CN103077347A (en) | Combined type intrusion detecting method on basis of data fusion of improved core vector machine | |
CN102263790A (en) | Intrusion detecting method based on integrated learning | |
CN104951764A (en) | Identification method for behaviors of high-speed vehicle based on secondary spectrum clustering and HMM (Hidden Markov Model)-RF (Random Forest) hybrid model | |
CN114612836B (en) | Monitoring video abnormity detection method based on memory-enhanced future video frame prediction | |
CN104007431A (en) | Radar HRRP target recognition method based on dpLVSVM model | |
CN102200787A (en) | Robot behaviour multi-level integrated learning method and robot behaviour multi-level integrated learning system | |
DE112020005663T5 (en) | OBJECT RECOGNITION WITH TRAINING FROM MULTIPLE DATASETS | |
Zhang et al. | Trajectory series analysis based event rule induction for visual surveillance | |
Chen et al. | Multi-granularity regularized re-balancing for class incremental learning | |
RU2708965C1 (en) | Method of analyzing functional behavior of a technical system and a processing unit | |
CN103295007B (en) | A kind of Feature Dimension Reduction optimization method for Chinese Character Recognition | |
Shah et al. | Software clustering using automated feature subset selection | |
CN102880881A (en) | Method for identifying car type on basis of binary support vector machines and genetic algorithm | |
CN109635009B (en) | Fuzzy matching inquiry system | |
Haq et al. | Automatic test suite generation for key-points detection dnns using many-objective search | |
CN109993336A (en) | Financial investment data mutation analysis method and its system based on wavelet analysis | |
Xu | Research on software credibility algorithm based on deep convolutional sparse coding | |
Saunier et al. | Investigating collison factors by mining microscopic data of vehicle conflicts and collisions | |
Zhu et al. | A robust feature fusion method for camera-based highway guardrail detection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request |
Effective date: 20150217 |
|
FZDE | Discontinued |
Effective date: 20160908 |
|
FZDE | Discontinued |
Effective date: 20160908 |