WO2004004313A1 - Image processing apparatus and method capable of finely adjusting data compression rate - Google Patents

Image processing apparatus and method capable of finely adjusting data compression rate Download PDF

Info

Publication number
WO2004004313A1
WO2004004313A1 PCT/JP2003/007034 JP0307034W WO2004004313A1 WO 2004004313 A1 WO2004004313 A1 WO 2004004313A1 JP 0307034 W JP0307034 W JP 0307034W WO 2004004313 A1 WO2004004313 A1 WO 2004004313A1
Authority
WO
WIPO (PCT)
Prior art keywords
image data
block
image processing
coding
frequency component
Prior art date
Application number
PCT/JP2003/007034
Other languages
French (fr)
Inventor
Yutaka Sato
Yukio Kadowaki
Kazuhiro Miyazaki
Original Assignee
Ricoh Company, Ltd.
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ricoh Company, Ltd. filed Critical Ricoh Company, Ltd.
Priority to EP03730806A priority Critical patent/EP1516482A4/en
Priority to US10/495,855 priority patent/US7574062B2/en
Publication of WO2004004313A1 publication Critical patent/WO2004004313A1/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/46—Colour picture communication systems
    • H04N1/64—Systems for the transmission or the storage of the colour picture signal; Details therefor, e.g. coding or decoding means therefor
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H04N19/126—Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/136—Incoming video signal characteristics or properties
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/18—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a set of transform coefficients
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
    • H04N19/635—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets characterised by filter definition or implementation details

Definitions

  • the present invention generally relates to image processing apparatuses and methods, and more particularly to an image processing apparatus and method performing coding based on image-related data divided into bit planes:
  • FIG. 1 is a block diagram showing a JPEG 2000-compliant image processing apparatus 50.
  • image data is converted to three signals of Y (luminance) , Cr (color difference) , and Cb (color difference) in a color conversion part 1.
  • a discrete wavelet transform (DWT) part 2 the three signals are separately subjected to discrete wavelet transform of level three using an irreversible transform filter called an IR 9/7 filter in JPEG 2000.
  • the wavelet coefficients obtained in the DWT part 2 are subjected to scalar quantization in a quantization part 3.
  • the wavelet coefficients are divided into blocks of a predetermined size in each sub-band so that the scalar-quantized data is decomposed, block by block, into bit planes in order from the MSBs (most significant bits) to the LSBs (least significant bits) .
  • FIG. 2 is a diagram showing a- state where the 8-bit wavelet coefficients of a block are decomposed into eight bit planes.
  • the coefficient modeling part 4 codes, or performs coefficient 'coding on, each of the wavelet coefficients decomposed into the bit planes through three separate coding- paths.
  • the coefficient coding a higher compression rate is achieved as the bit planes include more data values of , 0."
  • the coefficient coding is not performed on a zero bit plane or a bit plane whose data values are all "0s,” but is performed after the initial ' appearance of a bit plane including a significant data value of w l.” Therefore, in the state of FIG. 2, the bit planes of the high-order two bits are not subjected to coefficient coding, and only the bit planes of the low-order six " bits are subjected to coefficient coding.
  • An MQ-coder part 5 performs binary arithmetic coding on the image data subjected to the coefficient coding in the coefficient modeling part 4 , and outputs the resultant coded data to a code formation part 6.
  • the coefficient modeling part 4 and the MQ-coder part 5 form a coding part.
  • the code formation part 6 adds data such as information on the coding to a predetermined part of the coded data obtained in the MQ-coder part 5 , thereby forming and outputting a series of coded data called a code stream.
  • a more specific object of the present invention is to provide an image processing apparatus and method that performs coding on image data based on the result of decomposing the image data into bit planes block by block and can generate coded data at a compression rate more finely adjusted with a simple configuration.
  • the above objects of the present invention are achieved by an image processing apparatus including: a coding part that performs coding on image data based on a result of decomposing the image data into bit planes block by block; and a quantization part that reduces values of image data of a predetermined block to be compressed at an increased rate, the quantization part being provided at an earlier stage than the coding part .
  • the values of the data of a predetermined block to be compressed at an increased rate are reduced, so that finer adjustment can be performed on the compression rate compared with the case of discarding low-order bit data of the data of the predetermined block.
  • an image processing method including the steps of (a) performing coding on image data based on a result of decomposing the image data into bit planes block by block; and (b) performing quantization that reduces values of image data of a predetermined block to be compressed at an increased rate before the step (a) .
  • the values of the data of a predetermined block to be compressed at an increased rate are reduced, so that finer adjustment can be performed on the compression rate compared with the case of discarding low-order bit data of the data of the predetermined block.
  • FIG. 1 is a block diagram showing a conventional JPEG 2000-compliant image processing apparatus
  • FIG. 2 is a diagram showing a state where the 8-bit wavelet coefficients of a block are decomposed into eight bit planes;
  • FIG. 3 is a diagram showing a JPEG 2000-compliant image processing apparatus according to an embodiment of the present invention.
  • FIG. 4 is a diagram showing wavelet coefficients generated by level-3 discrete wavelet transform (DWT) according to the embodiment of the present invention
  • FIG. 5 is a diagram showing a quantization part of the image processing apparatus of FIG. 3 according to the embodiment of the present invention
  • FIG. 6 is a diagram showing a state where the wavelet coefficients of a block are divided into bit. planes according to the embodiment of the present invention.
  • FIG. 7 is a diagram showing "digit reduction" caused by reducing a wavelet coefficient by a proportion.
  • FIG. 3 is a diagram showing a JPEG 2000-compliant image processing apparatus 100 according to the embodiment of the present invention.
  • image data is converted to three signals of Y (luminance) , Cr (color difference) , and Cb (color difference) .
  • Y luminance
  • Cr color difference
  • Cb color difference
  • the signals are separately subjected to wavelet transform of level 3 using a reversible transform filter called an R 5/3 filter in JPEG 2000.
  • FIG. 4 is a diagram showing the wavelet coefficients generated by the level-3 DWT performed in the DWT part 7 using the reversible transform filter. It is known that of the wavelet coefficients, those of the blocks of low-frequency component sub-bands such as 3LL, 3LH, 3HL, and 3HH relate to the overall contour of a reproduced image and those of the blocks of high-frequency component sub-bands such as 1HH, 1LH, and 1HL relate to the resolution of a reproduced image . 5 Referring back to FIG. 3, compared with the irreversible transform filter (or IR 9/7 filter in JPEG 2000) , the reversible transform filter employed in the DWT part 7 has a simple configuration and can be reduced in size with ease. Therefore, the image processing apparatus 100 can be reduced
  • the quantization part 8 reduces the value of each wavelet coefficient by a proportion that becomes smaller for a block
  • the proportion (quantization step) is finer than 2 n .
  • the coefficient coding performed in the coefficient modeling part 4 at a later stage achieves a higher compression rate.
  • the reduction of the values of the wavelet coefficients of the high-frequency component blocks in the quantization part 8 takes advantage of the well-known characteristic that the reduction of the values of the wavelet coefficients of a high-frequency component block does not • affect the quality of a reproduced .image. Since the quantization part 8 is provided, the compression rate at the time of the coefficient coding performed, in the coefficient modeling, part 4 can be increased while the degradation of the quality of a reproduced image is prevented. Further, compared with the case of discarding the data of low-order bit planes, finer adjustment can be performed on the compression rate. In the coefficient modeling part 4, the wavelet
  • coefficients are divided into blocks each of a 16 X 16 pixel matrix in each sub-band, and the quantized data is decomposed, block by block, into bit planes in order from the MSBs to LSBs
  • the coefficient modeling part 4 codes , or performs coefficient coding on, each value of the image data decomposed into the bit planes through three separate coding paths .
  • the coefficient coding In the coefficient coding, a higher compression rate is achieved as the bit plane includes more data values of "0.” Further, the coefficient coding is not performed on a zero bit plane or a bit plane whose data values are all "0s,” but is performed after the initial appearance of a bit plane including a significant data value of "1.” That is, in the coefficient coding, the compression rate is simply increased as the number of zero bit planes is increased.
  • the MQ-coder part 5 performs binary arithmetic coding on the image data subjected to the coefficient coding, and outputs the resultant coded data to the code formation part 6.
  • the coefficient modeling part 4 and the MQ-coder part 5 form a coding part .
  • the code formation part 6 adds data such as information on the coding to a predetermined part of the coded data obtained in the MQ-coder part 5 , thereby forming and • outputting a series of coded data called a code stream.
  • FIG. 5 is a diagram showing a configuration of the quantization part 8.
  • the values of the wavelet coefficients are reduced by a proportion that increases so as to be the smallest for the low-frequency component blocks (3LL, 3LH, 3HL, and 3HH) and the largest for the high-frequency component blocks (1LH, 1HL, 1HH) .
  • the wavelet coefficients input to the quantization part 8 are input to a counter 10 and a multiplier 14.
  • the count value of the counter 10 is input to a comparator 11 via one of a pair of the terminals thereof and to a comparator 12 via one of a pair of the terminals thereof.
  • the other terminal of the comparator 11 is set to ref 1 corresponding to the number of wavelet coefficients, that is, the number of pixels, of the 3LL, 3LH, 3HL, and 3HH blocks (16
  • the other terminal of the comparator 12 is set to ref 2 corresponding to the number of wavelet coefficients, that is, the number of pixels, of the 2LL (3LL, 3LH, 3HL, and 3HH) , 2LH, 2HL, and 2HH blocks (1024 +
  • the level of the output signal of the comparator 11 switches from LOW to HIGH.
  • the level of the output signal of the comparator 12 switches from LOW to HIGH.
  • the output signals of the comparators 11 and 12 are input to a selector 13 as the low-order bit signal and the high-order bit signal, respectively, of a two-bit selection signal.
  • the selector 13 outputs a multiplier of " l.O to the multiplier 14. In the case of "01,” the selector 13 outputs a multiplier of 0.9 to the multiplier 14. In the case of "11,” the selector 13 outputs a multiplier of 0.6 to the multiplier 14.
  • the wavelet coefficients of the low- frequency component blocks (3LL, 3LH, 3HL, and 3HH) are output directly to the coefficient modeling part 4, the wavelet coefficients of the blocks of the middle frequency range (2LH, 2HL, and 2HH) are output to the coefficient modeling part 4 after being multiplied by 0.9, and the wavelet coefficients of the high-frequency component blocks (1LH, 1HL, and 1HH) are output to the coefficient modeling part 4 after being multiplied by 0.6.
  • the coefficient coding is not performed on a zero bit plane whose data values are all "0s"' but is performed after the initial appearance of a bit plane including a significant data value of ⁇ , l,"
  • the wavelet coefficients of the high-frequency component blocks (1LH, 1HL, and 1HH) may be reduced in value by one or more digits by multiplying the wavelet coefficients by a value less than or equal to 0.5 so that an increased number of zero bit planes are formed in succession from the MSB plane toward a lower- order bit plane. Consequently, the coefficient coding can be performed on the high-frequency component blocks with more efficiency.
  • FIG. 6 is a diagram .showing a state where the
  • quantized wavelet coefficients of the 3LL block of a 16 16 pixel matrix are divided into bit planes in order from their MSBs to LSBs. As shown in FIG. 6, in the case of 8-bit wavelet coefficients , the wavelet coefficients are divided into eight bit planes .
  • FIG. 7 is a diagram showing the case of reducing a wavelet coefficient of a ILL block in value by a digit by multiplying the coefficient by 0.6 in the quantization part 8.
  • "10111001” multiplied by 0.6 is "01110000.”
  • "digit reduction” occurs so that the value of the ' MSB of the coefficient changes from "1" to "0.”
  • the coefficient modeling part 4 performs coefficient coding on each value of the image data decomposed into the bit planes through three separate coding paths. At ' this point, the compression rate can be increased as the bit planes include more data values of "0.” Further, the .
  • coefficient coding is not performed on a zero bit plane whose data values are all "0s" but is performed after the initial appearance of a bit " plane including a significant data value of "1.” That is, by particularly reducing the wavelet coefficients of high-frequency component blocks by a large proportion, the number of zero bit planes is increased, so that the compression rate is increased.
  • the image processing apparatus 100 of this embodiment which employs the DWT part 7 using a reversible transform filter simple in configuration and reducible in size, reduces the values of the wavelet coefficients of high-frequency component blocks having little effect on the quality of a reproduced image in the quantization part 8. Consequently, the compression rate of coded data is increased, and finer adjustment can.be performed on the compression rate than in the case of adjusting the compression rate by discarding ' the data of low-order bits of the wavelet coefficients .
  • the image processing apparatus 100 of- this embodiment is applicable to apparatuses including a digital still camera, the recording part of a digital copier, a monitor camera, a digital video storage device, and a digital' camera (for moving pictures) .

Abstract

ABSTRACT An image processing apparatus includes a coding part that performs coding on image data based on the result of decomposing the image data into bit planes block by block, and a quantization part that reduces the values of the image data of a predetermined block to be compressed at an increased rate. The quantization part is provided at an earlier stage than the coding part.

Description

DESCRIPTION
IMAGE PROCESSING APPARATUS AND METHOD CAPABLE OF FINELY ADJUSTING DATA COMPRESSION RATE
TECHNICAL. FIELD
The present invention generally relates to image processing apparatuses and methods, and more particularly to an image processing apparatus and method performing coding based on image-related data divided into bit planes:
BACKGROUND ART
There are conventionally known image processing apparatuses that perform coding based on quantized image data divided into bit planes. Such image processing apparatuses include those in compliance with JPEG 2000, for instance. FIG. 1 is a block diagram showing a JPEG 2000-compliant image processing apparatus 50. In the image processing apparatus 50, first, image data is converted to three signals of Y (luminance) , Cr (color difference) , and Cb (color difference) in a color conversion part 1. In a discrete wavelet transform (DWT) part 2, the three signals are separately subjected to discrete wavelet transform of level three using an irreversible transform filter called an IR 9/7 filter in JPEG 2000. The wavelet coefficients obtained in the DWT part 2 are subjected to scalar quantization in a quantization part 3.
Next, in a coefficient modeling part 4, the wavelet coefficients are divided into blocks of a predetermined size in each sub-band so that the scalar-quantized data is decomposed, block by block, into bit planes in order from the MSBs (most significant bits) to the LSBs (least significant bits) .
FIG. 2 is a diagram showing a- state where the 8-bit wavelet coefficients of a block are decomposed into eight bit planes. The coefficient modeling part 4 codes, or performs coefficient 'coding on, each of the wavelet coefficients decomposed into the bit planes through three separate coding- paths. In the coefficient coding, a higher compression rate is achieved as the bit planes include more data values of ,0." Further, the coefficient coding is not performed on a zero bit plane or a bit plane whose data values are all "0s," but is performed after the initial ' appearance of a bit plane including a significant data value of wl." Therefore, in the state of FIG. 2, the bit planes of the high-order two bits are not subjected to coefficient coding, and only the bit planes of the low-order six "bits are subjected to coefficient coding.
An MQ-coder part 5 performs binary arithmetic coding on the image data subjected to the coefficient coding in the coefficient modeling part 4 , and outputs the resultant coded data to a code formation part 6. In the image processing apparatus 50, the coefficient modeling part 4 and the MQ-coder part 5 form a coding part.
The code formation part 6 adds data such as information on the coding to a predetermined part of the coded data obtained in the MQ-coder part 5 , thereby forming and outputting a series of coded data called a code stream.
DISCLOSURE OF THE INVENTION According to the JPEG 2000-compliant coding, in order to achieve a higher compression rate, the data of low- order bit planes of the wavelet coefficients subjected to the coefficient coding of the coefficient modeling part 4 is discarded. According to this method, however, the data values can be changed only by 2n, so that the compression rate can be adjusted only roughly.
Accordingly, it is a general object of the present invention to provide an image processing apparatus and method in which the above-described disadvantage is eliminated. A more specific object of the present invention is to provide an image processing apparatus and method that performs coding on image data based on the result of decomposing the image data into bit planes block by block and can generate coded data at a compression rate more finely adjusted with a simple configuration. The above objects of the present invention are achieved by an image processing apparatus including: a coding part that performs coding on image data based on a result of decomposing the image data into bit planes block by block; and a quantization part that reduces values of image data of a predetermined block to be compressed at an increased rate, the quantization part being provided at an earlier stage than the coding part .
According to the above-described image processing apparatus, of the values of image data before being subjected to coding, the values of the data of a predetermined block to be compressed at an increased rate are reduced, so that finer adjustment can be performed on the compression rate compared with the case of discarding low-order bit data of the data of the predetermined block.
The above objects of the present invention are also achieved by an image processing method including the steps of (a) performing coding on image data based on a result of decomposing the image data into bit planes block by block; and (b) performing quantization that reduces values of image data of a predetermined block to be compressed at an increased rate before the step (a) .
According to the above described image processing method, of the values of image data before being subjected to coding, the values of the data of a predetermined block to be compressed at an increased rate are reduced, so that finer adjustment can be performed on the compression rate compared with the case of discarding low-order bit data of the data of the predetermined block.
BRIEF DESCRIPTION OF THE DRAWINGS
Other objects, features and advantages of the present invention will become more apparent from the following detailed description when read in conjunction with the accompanying drawings, in which:
FIG. 1 is a block diagram showing a conventional JPEG 2000-compliant image processing apparatus;
FIG. 2 is a diagram showing a state where the 8-bit wavelet coefficients of a block are decomposed into eight bit planes;
FIG. 3 is a diagram showing a JPEG 2000-compliant image processing apparatus according to an embodiment of the present invention;
FIG. 4 is a diagram showing wavelet coefficients generated by level-3 discrete wavelet transform (DWT) according to the embodiment of the present invention;
FIG. 5 is a diagram showing a quantization part of the image processing apparatus of FIG. 3 according to the embodiment of the present invention; FIG. 6 is a diagram showing a state where the wavelet coefficients of a block are divided into bit. planes according to the embodiment of the present invention; and
FIG. 7 is a diagram showing "digit reduction" caused by reducing a wavelet coefficient by a proportion.
BEST MODE FOR CARRYING OUT THE INVENTION (Entire configuration)
A description will now be given, with reference to the accompanying drawings, of an embodiment of the present invention. FIG. 3 is a diagram showing a JPEG 2000-compliant image processing apparatus 100 according to the embodiment of the present invention. The same elements as those of the above-described image processing apparatus 50 are referred to by the same numerals . Initially, in the color conversion part of the image processing apparatus 100, image data is converted to three signals of Y (luminance) , Cr (color difference) , and Cb (color difference) . Then, in a DWT part 7, the signals are separately subjected to wavelet transform of level 3 using a reversible transform filter called an R 5/3 filter in JPEG 2000.
FIG. 4 is a diagram showing the wavelet coefficients generated by the level-3 DWT performed in the DWT part 7 using the reversible transform filter. It is known that of the wavelet coefficients, those of the blocks of low-frequency component sub-bands such as 3LL, 3LH, 3HL, and 3HH relate to the overall contour of a reproduced image and those of the blocks of high-frequency component sub-bands such as 1HH, 1LH, and 1HL relate to the resolution of a reproduced image . 5 Referring back to FIG. 3, compared with the irreversible transform filter (or IR 9/7 filter in JPEG 2000) , the reversible transform filter employed in the DWT part 7 has a simple configuration and can be reduced in size with ease. Therefore, the image processing apparatus 100 can be reduced
10 in size. In the case of employing a reversible transform filter, however, the compression rate is low. Therefore, according to the image processing apparatus 100, the values of the wavelet coefficients of the high-frequency component blocks that have little effect on the quality of a reproduced
15 image are reduced in a below-described quantization part 8 so as to increase the compression rate.
As will be later described in detail, the quantization part 8 reduces the value of each wavelet coefficient by a proportion that becomes smaller for a block
20. of a lower-frequency component and larger for a block of a higher-frequency component. Particularly, the proportion (quantization step) is finer than 2n. By this processing, some wavelet coefficients become smaller in value by one or more digits, thereby increasing data values of "0" existing in the
25 bit planes. As a result, the coefficient coding performed in the coefficient modeling part 4 at a later stage achieves a higher compression rate.
The reduction of the values of the wavelet coefficients of the high-frequency component blocks in the quantization part 8 takes advantage of the well-known characteristic that the reduction of the values of the wavelet coefficients of a high-frequency component block does not • affect the quality of a reproduced .image. Since the quantization part 8 is provided, the compression rate at the time of the coefficient coding performed, in the coefficient modeling, part 4 can be increased while the degradation of the quality of a reproduced image is prevented. Further, compared with the case of discarding the data of low-order bit planes, finer adjustment can be performed on the compression rate. In the coefficient modeling part 4, the wavelet
coefficients are divided into blocks each of a 16 X 16 pixel matrix in each sub-band, and the quantized data is decomposed, block by block, into bit planes in order from the MSBs to LSBs The coefficient modeling part 4 codes , or performs coefficient coding on, each value of the image data decomposed into the bit planes through three separate coding paths . In the coefficient coding, a higher compression rate is achieved as the bit plane includes more data values of "0." Further, the coefficient coding is not performed on a zero bit plane or a bit plane whose data values are all "0s," but is performed after the initial appearance of a bit plane including a significant data value of "1." That is, in the coefficient coding, the compression rate is simply increased as the number of zero bit planes is increased. The MQ-coder part 5 performs binary arithmetic coding on the image data subjected to the coefficient coding, and outputs the resultant coded data to the code formation part 6.
In the image processing apparatus 100, the coefficient modeling part 4 and the MQ-coder part 5 form a coding part .
The code formation part 6 adds data such as information on the coding to a predetermined part of the coded data obtained in the MQ-coder part 5 , thereby forming and • outputting a series of coded data called a code stream.
(Configuration of the quantization part 8) FIG. 5 is a diagram showing a configuration of the quantization part 8. In the quantization part 8, the values of the wavelet coefficients are reduced by a proportion that increases so as to be the smallest for the low-frequency component blocks (3LL, 3LH, 3HL, and 3HH) and the largest for the high-frequency component blocks (1LH, 1HL, 1HH) . The wavelet coefficients input to the quantization part 8 are input to a counter 10 and a multiplier 14. The count value of the counter 10 is input to a comparator 11 via one of a pair of the terminals thereof and to a comparator 12 via one of a pair of the terminals thereof. The other terminal of the comparator 11 is set to ref 1 corresponding to the number of wavelet coefficients, that is, the number of pixels, of the 3LL, 3LH, 3HL, and 3HH blocks (16
X 16 X 4 = 1024). Further, the other terminal of the comparator 12 is set to ref 2 corresponding to the number of wavelet coefficients, that is, the number of pixels, of the 2LL (3LL, 3LH, 3HL, and 3HH) , 2LH, 2HL, and 2HH blocks (1024 +
32 X 32 X 3 = 4096) .
When the count value of the counter 10 reaches ref-1, the level of the output signal of the comparator 11 switches from LOW to HIGH. When the count value of the counter 10 reaches ref 2, the level of the output signal of the comparator 12 switches from LOW to HIGH. The output signals of the comparators 11 and 12 are input to a selector 13 as the low-order bit signal and the high-order bit signal, respectively, of a two-bit selection signal. When the input two-bit selection signal is "00," the selector 13 outputs a multiplier of "l.O to the multiplier 14. In the case of "01," the selector 13 outputs a multiplier of 0.9 to the multiplier 14. In the case of "11," the selector 13 outputs a multiplier of 0.6 to the multiplier 14. According to the quantization part 8 of the above- described configuration, the wavelet coefficients of the low- frequency component blocks (3LL, 3LH, 3HL, and 3HH) are output directly to the coefficient modeling part 4, the wavelet coefficients of the blocks of the middle frequency range (2LH, 2HL, and 2HH) are output to the coefficient modeling part 4 after being multiplied by 0.9, and the wavelet coefficients of the high-frequency component blocks (1LH, 1HL, and 1HH) are output to the coefficient modeling part 4 after being multiplied by 0.6. When it is taken into consideration that in the coefficient modeling part 4 , the coefficient coding is not performed on a zero bit plane whose data values are all "0s"' but is performed after the initial appearance of a bit plane including a significant data value of Λ,l," the wavelet coefficients of the high-frequency component blocks (1LH, 1HL, and 1HH) may be reduced in value by one or more digits by multiplying the wavelet coefficients by a value less than or equal to 0.5 so that an increased number of zero bit planes are formed in succession from the MSB plane toward a lower- order bit plane. Consequently, the coefficient coding can be performed on the high-frequency component blocks with more efficiency.
FIG. 6 is a diagram .showing a state where the
quantized wavelet coefficients of the 3LL block of a 16 16 pixel matrix are divided into bit planes in order from their MSBs to LSBs. As shown in FIG. 6, in the case of 8-bit wavelet coefficients , the wavelet coefficients are divided into eight bit planes .
FIG. 7 is a diagram showing the case of reducing a wavelet coefficient of a ILL block in value by a digit by multiplying the coefficient by 0.6 in the quantization part 8. In this case, "10111001" multiplied by 0.6 is "01110000." As a result, "digit reduction" occurs so that the value of the ' MSB of the coefficient changes from "1" to "0." As previously described, the coefficient modeling part 4 performs coefficient coding on each value of the image data decomposed into the bit planes through three separate coding paths. At' this point, the compression rate can be increased as the bit planes include more data values of "0." Further, the. coefficient coding is not performed on a zero bit plane whose data values are all "0s" but is performed after the initial appearance of a bit" plane including a significant data value of "1." That is, by particularly reducing the wavelet coefficients of high-frequency component blocks by a large proportion, the number of zero bit planes is increased, so that the compression rate is increased.
As described above, the image processing apparatus 100 of this embodiment, which employs the DWT part 7 using a reversible transform filter simple in configuration and reducible in size, reduces the values of the wavelet coefficients of high-frequency component blocks having little effect on the quality of a reproduced image in the quantization part 8. Consequently, the compression rate of coded data is increased, and finer adjustment can.be performed on the compression rate than in the case of adjusting the compression rate by discarding ' the data of low-order bits of the wavelet coefficients .
The image processing apparatus 100 of- this embodiment is applicable to apparatuses including a digital still camera, the recording part of a digital copier, a monitor camera, a digital video storage device, and a digital' camera (for moving pictures) .
The present invention is not limited to the specifically disclosed embodiment, but variations and modifications may be made without departing from the scope of the present invention.
The present application is based on Japanese priority application No. 2002-187543 filed on June 27, 2002, the entire contents of which are hereby incorporated by reference.

Claims

1. An image processing apparatus, comprising: a coding part that performs coding on image data based on a result of decomposing the image data into bit planes block by block; and a quantization part that reduces values of image data of a predetermined block to be compressed at an increased rate, the quantization part being provided at an earlier stage than the coding part.
2. The image processing apparatus as claimed in claim 1, wherein: said quantization part reduces the values of the image data of the predetermined block so that bit planes of the values of the image data of the predetermined block are generated with an increased number of zero bit planes whose data values are all binary 0s being formed in succession from a most-significant-bit plane toward a lower-order bit plane; and the coding part performs the coding on the bit • planes of the image data of the predetermined block except for the zero bit planes .
3. The image processing apparatus as claimed in claim 1 , wherein of wavelet coefficients of sub-bands obtained by performing discrete wavelet transform on the image data, said quantization part reduces wavelet coefficients of a block of a high-frequency component in value.
4. The image processing apparatus as claimed in claim 1, wherein said quantization part reduces wavelet coefficients of sub-bands in value by a proportion that increases ' for a block of a .higher-frequency component and decreases for a block of a lower-frequency component, the sub- bands being obtained by performing discrete wavelet transform on the image data .
5. The image processing apparatus as claimed in claim 1,. further comprising: a discrete wavelet transform part that performs discrete wavelet transform on the image data using a reversible transform filter, wherein : the image processing apparatus is JPEG 2000- compliant; said quantization part reduces wavelet coefficients of sub-bands in value by a proportion that increases for a block of a higher-frequency component and decreases, for a block of a ■ lower-frequency component, the sub-bands being obtained by the discrete wavelet transform on the image data
by said discrete wavelet transform part; and
the coding part performs JPEG 2000-compliant' entropy coding on- the wavelet coefficients quantized in said
■ 5 quantization part.
6. An image processing method, comprising the steps
of:
(a) performing coding on image data based on a
10 result of decomposing the image data into bit planes block by
block; and
(b) performing quantization that reduces values of image data of a predetermined block to be compressed at an
increased rate before said step (a) .
15
7. The image processing method as claimed in claim
6 , wherein : said step (b) reduces the values of the image data
of the predetermined block so that bit planes of the values of 20 the image data of the predetermined block are generated with an increased number of zero bit planes whose data values are all binary 0s being formed in succession from a most-
significant-bit plane toward a lower-order bit plane; and said step (a) performs the coding on the bit planes 25 of the image data of the . predetermined block except for the zero bit planes .
8. The image processing method as claimed in claim 6 , wherein of wavelet coefficients of sub-bands obtained by performing discrete wavelet transform on the image data, said step (b) reduces wavelet coefficients .of a block of a high- frequency component in value .
9. The image processing method as claimed in claim 6, wherein said step (b) reduces wavelet coefficients of sub- bands in value by a proportion that increases for a block of a higher-frequency component and decreases for a block of a lower-frequency component, the sub-bands being obtained by performing discrete wavelet transform on the image data.
10. The image processing method as claimed in claim 6, further comprising the step of (c) performing discrete wavelet transform on the image data using a reversible transform filter, wherein: the image processing method performs JPEG 2000- compliant coding; said step (b) reduces wavelet coefficients of sub- bands in value by a proportion that increases for a block of a higher-frequency component and decreases for a block of a lower-frequency component, the sub-bands being obtained by the discrete wavelet transform performed on the image data by said step (c) ; and said step (a) performs JPEG 2000-compliant entropy coding on the wavelet coefficients obtained after said step ■ (b) .
PCT/JP2003/007034 2002-06-27 2003-06-03 Image processing apparatus and method capable of finely adjusting data compression rate WO2004004313A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP03730806A EP1516482A4 (en) 2002-06-27 2003-06-03 Image processing apparatus and method capable of finely adjusting data compression rate
US10/495,855 US7574062B2 (en) 2002-06-27 2003-06-03 Image processing apparatus and method capable of finely adjusting data compression rate

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2002187543A JP2004032493A (en) 2002-06-27 2002-06-27 Image processing apparatus and image processing method
JP2002-187543 2002-06-27

Publications (1)

Publication Number Publication Date
WO2004004313A1 true WO2004004313A1 (en) 2004-01-08

Family

ID=29996790

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2003/007034 WO2004004313A1 (en) 2002-06-27 2003-06-03 Image processing apparatus and method capable of finely adjusting data compression rate

Country Status (4)

Country Link
US (1) US7574062B2 (en)
EP (1) EP1516482A4 (en)
JP (1) JP2004032493A (en)
WO (1) WO2004004313A1 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008228208A (en) * 2007-03-15 2008-09-25 Ricoh Co Ltd Image encoding method, image encoding device, image encoding circuit, information recording medium, and computer program
US20110176743A1 (en) * 2010-01-21 2011-07-21 Sankar Pathamadi V Data compression methods

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH08149308A (en) * 1994-11-22 1996-06-07 Ricoh Co Ltd Image coder, image coder decoder and image coding and decoding method
JP2002057903A (en) * 2000-05-26 2002-02-22 Eastman Kodak Co Method for generating compressed digital image organized into layer with information relating to different view condition and resolution
JP2002135781A (en) * 2000-10-25 2002-05-10 Canon Inc Image encoding apparatus, image decoding apparatus methods for them and recording medium
JP2002369202A (en) * 2001-06-11 2002-12-20 Ricoh Co Ltd Image compression device, image expansion device, image compressing method, image expanding method, program and recording medium stored with the program

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3209171B2 (en) * 1998-01-08 2001-09-17 日本電気株式会社 Video encoding device
US7369161B2 (en) * 1999-06-08 2008-05-06 Lightsurf Technologies, Inc. Digital camera device providing improved methodology for rapidly taking successive pictures
JP2001298623A (en) 2000-04-14 2001-10-26 Canon Inc Encoder and encoding method
US6408026B1 (en) * 1999-08-06 2002-06-18 Sony Corporation Deadzone quantization method and apparatus for image compression
US6944350B2 (en) * 1999-12-17 2005-09-13 Utah State University Method for image coding by rate-distortion adaptive zerotree-based residual vector quantization and system for effecting same
JP2001231009A (en) 2000-02-18 2001-08-24 Canon Inc Image data storage device ad method
US7035473B1 (en) * 2000-03-01 2006-04-25 Sharp Laboratories Of America, Inc. Distortion-adaptive visual frequency weighting
US6985632B2 (en) * 2000-04-17 2006-01-10 Canon Kabushiki Kaisha Image processing system, image processing apparatus, and image processing method
JP3702778B2 (en) 2000-11-27 2005-10-05 ソニー株式会社 Image coding apparatus and method
US6492916B1 (en) * 2001-03-30 2002-12-10 Ricoh Co., Ltd. Method and apparatus for generating multiple selectable contexts
US7155066B2 (en) * 2001-05-31 2006-12-26 Agilent Technologies, Inc. System and method for demosaicing raw data images with compression considerations
US7031494B2 (en) * 2002-06-24 2006-04-18 Eastman Kodak Company Method for securely transacting a transaction based on a transaction document

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH08149308A (en) * 1994-11-22 1996-06-07 Ricoh Co Ltd Image coder, image coder decoder and image coding and decoding method
JP2002057903A (en) * 2000-05-26 2002-02-22 Eastman Kodak Co Method for generating compressed digital image organized into layer with information relating to different view condition and resolution
JP2002135781A (en) * 2000-10-25 2002-05-10 Canon Inc Image encoding apparatus, image decoding apparatus methods for them and recording medium
JP2002369202A (en) * 2001-06-11 2002-12-20 Ricoh Co Ltd Image compression device, image expansion device, image compressing method, image expanding method, program and recording medium stored with the program

Also Published As

Publication number Publication date
US7574062B2 (en) 2009-08-11
US20040264783A1 (en) 2004-12-30
JP2004032493A (en) 2004-01-29
EP1516482A4 (en) 2007-08-01
EP1516482A1 (en) 2005-03-23

Similar Documents

Publication Publication Date Title
US6301392B1 (en) Efficient methodology to select the quantization threshold parameters in a DWT-based image compression scheme in order to score a predefined minimum number of images into a fixed size secondary storage
US8000542B2 (en) Apparatus and method for processing image data based on object movement speed within a frame
JP2003319185A (en) Image encoding apparatus and method, image decoding apparatus and method
JP2002520899A (en) Real-time algorithm and architecture for encoding images compressed by DWT-based techniques
JP2000115770A (en) Coder and its method
US5737448A (en) Method and apparatus for low bit rate image compression
US7120306B2 (en) Image processing method and image coding apparatus utilizing the image processing method
JP2010093669A (en) Apparatus and method for processing information
JP2008514142A (en) Multi-technology entropy coding system and method
US7525463B2 (en) Compression rate control system and method with variable subband processing
US20110091123A1 (en) Coding apparatus and coding method
US7224846B2 (en) Image coding apparatus and method, and program and recording medium
KR20100079121A (en) Image encoder and image encoding method
JP2004040429A (en) Digital image encoder, digital image encoding method used therefor, and program therefor
US20040131271A1 (en) Image compression method using wavelet transform
JP2014521273A (en) Method and apparatus for encoding an image
WO2004004313A1 (en) Image processing apparatus and method capable of finely adjusting data compression rate
US20060072834A1 (en) Permutation procrastination
EP1609303B1 (en) Image compression device and method
US7440630B2 (en) Image compressing apparatus that achieves desired code amount
US8279098B2 (en) Compression rate control system and method with variable subband processing
JP4124436B2 (en) Motion estimation device, program, storage medium, and motion estimation method
US20090074059A1 (en) Encoding method and device for image data
JP2004048212A (en) Digital image coder, coding method and program
JP2007184953A (en) Image processing apparatus

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT SE SI SK TR

DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 10495855

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 2003730806

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2003730806

Country of ref document: EP