CN100491906C - Grating projection and recognition method - Google Patents
Grating projection and recognition method Download PDFInfo
- Publication number
- CN100491906C CN100491906C CNB2005100958914A CN200510095891A CN100491906C CN 100491906 C CN100491906 C CN 100491906C CN B2005100958914 A CNB2005100958914 A CN B2005100958914A CN 200510095891 A CN200510095891 A CN 200510095891A CN 100491906 C CN100491906 C CN 100491906C
- Authority
- CN
- China
- Prior art keywords
- sub
- gratings
- striation
- segmentation
- grating
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The related improvement on high-density grating projection and recognition method comprises: applying 'pyramid sub-grating projection method' to refine the original grating, then recognizing the optical strip of the treated grating. This invention can implement grating projection of arbitrary density and mode, and can avoid error led by interpolation.
Description
Technical field
The invention belongs to structural light three-dimensional topography measurement technology, relate to the improvement of high dencity grating projection with recognition methods.
Background technology
In the three dimensional shape measurement system, the custom-designed hardware system of the many employings of the pattern projector often can only throw the pattern of certain pattern.In the existing system, the single striation of the projection that has is realized measuring three-dimensional morphology [Zhu Zhou etc. by scanning, based on the three-dimensional whole body body-scanner of structured light, Central China University of Science and Technology's journal, 2004, Vol:32 (10): 7-9], [Feng Zhanjun etc. are based on the three-dimensional active vision detection of grating, measurement technology for the raster pattern that is formed by many striations that then throws that has, 2003,5:15-18].In the actual measurement, the density of grating striation is big more, just can obtain the three-dimensional appearance data of object more more meticulously.But when striation density was too big, difficulty just took place in the identification of striation, often was difficult to distinguish.Therefore, the grating that divides that the general at present method that adopts is the certain density of projection obtains more data by interpolation.Interpolation is a kind of approximate, must introduce approximate error.
Summary of the invention
The objective of the invention is: propose a kind of method of throwing high dencity grating and being easy to carry out striation identification, thereby improve the fineness of measuring three-dimensional morphology, the approximate error of avoiding interpolation to bring.
Technical scheme of the present invention is: a kind of grating projection and recognition methods, use digital projector as the raster pattern projector, digital projector is connected with a computing machine, produce the grating of arbitrary patterns by the computer control raster pattern projector, determine the density of the original grating that throws according to measuring needs, it is characterized in that
1, adopt " pyramid sub-gratings projective method " that raster density is reached the original grating that striation can not resolution degree and segment, establishing original grating has N root striation, N=2
n, n is a positive integer, the concrete steps of segmentation are:
1.1, every a striation original grating is subdivided into two width of cloth sub-gratings, every width of cloth sub-gratings contains N/2 root striation, every striation invariant position in original grating, the segmentation back is if sub-gratings reaches to differentiate to be required promptly to stop segmentation, otherwise proceeds next step segmentation;
1.2, to the segmentation after two width of cloth sub-gratings, continue to adopt the method for 1.1.1 to segment, original grating is subdivided into four width of cloth sub-gratings, every width of cloth sub-gratings contains N/4 root striation, the segmentation back requires promptly to stop segmentation if sub-gratings reaches to differentiate, otherwise proceed next step segmentation, requirement promptly stops segmentation if sub-gratings reaches resolution in the segmentation back, otherwise proceeds next step segmentation;
1.3, repeat the fine division step of 1.1.1, till the striation in the sub-gratings after segmentation can be differentiated identification;
2, the striation in the every width of cloth sub-gratings after the segmentation is discerned;
Adopt the two fens compiling methods in space to carry out striation identification, concrete steps are:
2.1, determine the striation number in the sub-gratings;
If original grating striation number is N, the segmentation number is M, and then the striation number in the sub-gratings of segmentation back is: N
Sub=N/M;
2.2, projection encode grating pattern;
For any width of cloth sub-gratings, every N
Sub/ 2, N
Sub/ 4 ..., up to throw a width of cloth encode grating pattern every 1 striation, need projection 1+log altogether
2(N
Sub) width of cloth coding pattern;
2.3, extract the striation center in the pre-treatment sub-gratings;
Adopt the method for history Dege to extract the striation center of working as in the pre-treatment sub-gratings, and calculate the coding at striation center, finish the coding of all striations in the pre-treatment sub-gratings according to the coding pattern;
2.4, repeating step 1.2.2 and 1.2.3, up to the identification of finishing all segmentation back sub-gratings striations, thereby finish the identification of original high dencity grating striation.
Advantage of the present invention is:
1, the inventive method can realize the projection of any density and pattern grating, and is very flexible, can realize the extremely careful description of object dimensional pattern, the approximate error of avoiding interpolation to bring.
2, under the high situation of raster density, the strategy by segmentation makes the grating striation become sparse and easy identification can improve the measuring three-dimensional morphology precision to a certain extent.
Description of drawings
Fig. 1 is the original grating synoptic diagram of high density.
Fig. 2 is a pyramid sub-gratings projective method synoptic diagram of the present invention.
Fig. 3 is the original grating pattern with 128 striations.
Fig. 4 and Fig. 5 are that Fig. 3 segments later sub-gratings pattern through the first time, and each sub-gratings has 64 striations.
Fig. 6 is that Fig. 3 segments later sub-gratings pattern through the second time, and each sub-gratings has 32 striations.
Fig. 7 be Fig. 3 through segmenting later sub-gratings pattern for the third time, each sub-gratings has 16 striations.
Fig. 8 is that Fig. 3 segments later sub-gratings pattern through the 4th time, and each sub-gratings has 8 striations.
Fig. 9 is that Fig. 3 segments later sub-gratings pattern through the 5th time, and each sub-gratings has 4 striations.
Figure 10 is that Fig. 3 segments later sub-gratings pattern through the 6th time, and each sub-gratings has 2 striations.
Figure 11 is that Fig. 3 segments later sub-gratings pattern through the 7th time, and each sub-gratings has 1 striation.
Embodiment
Below the present invention is described in further details.The present invention uses the very ripe digital projector of present technical conditions as the raster pattern projector.Projector is connected on the computing machine, can produce the grating of arbitrary patterns and any density by computing machine.Fig. 1 is the raster pattern of width of cloth projection.Raster pattern as shown in Figure 1, striation wherein is very dense, but can also differentiate identification.If raster density reaches the unresolvable degree of striation, then prior art just can't be carried out striation identification.If N (N=2 is arranged
n, n is a positive integer) and the root striation, the present invention adopts following divided method to solve this problem, is called " pyramid sub-gratings projective method ".
The present invention adopts " pyramid sub-gratings projective method " that raster density is reached the original grating that striation can not resolution degree and segments, and step is as follows:
(1) be two width of cloth sub-gratings every a striation with original grating subdivision at first, every width of cloth sub-gratings contains N/2 root striation, the invariant position of every striation in original grating, and promptly two width of cloth sub-gratings are combined as original raster pattern to folding.
(2) if after being subdivided into two width of cloth sub-gratings, the striation in the sub-gratings still reaches the density that can not differentiate or be difficult to differentiate, then can continue to adopt the method for (1) to segment to two width of cloth sub-gratings.Like this, original grating is subdivided into four width of cloth sub-gratings, and every width of cloth sub-gratings contains N/4 root striation, and the striation density in so grating further reduces.
(3) repeating step (1), the striation in the sub-gratings after segmentation can be differentiated identification.
" pyramid sub-gratings projective method " is an intensive grating subdivision M=2 in fact
n(n is a positive integer) width of cloth sub-gratings throws.
Be that example is described " pyramid sub-gratings projective method " with Fig. 2 below, original grating has 16 striations among Fig. 2.Become two width of cloth sub-gratings through segmentation for the first time, every width of cloth sub-gratings has 8 striations.Through segmentation for the second time, every width of cloth sub-gratings has 4 striations.Like this, can be subdivided into the sub-gratings that striation can be discerned to original grating.
On the basis of segmentation original grating, the present invention adopts the two fens compiling method [Tao Guozhi in space, the Zhu, a state in the Zhou Dynasty is continued expensive, the leaf good reputation, " coding method of raster pattern structured light sensor ", the accurate manufacturing technology of aviation, Vol.35, No.3,1999, pp.18-20.], the striation in the every width of cloth sub-gratings after the segmentation is discerned.The step of identification is:
1. suppose that original grating striation number is N, the segmentation number is M, and then the striation number in the sub-gratings of segmentation back is:
N
sub＝N/M
2. for any width of cloth sub-gratings, every N
Sub/ 2, N
Sub/ 4 ..., up to throw a width of cloth encode grating pattern every 1 striation, need projection 1+log altogether
2(N
Sub) width of cloth coding pattern.
3. adopt the history Dege (Ka Sideng. the history Dege, a kind of no inclined to one side detector of curvilinear structures, international Institute of Electrical and Electric Engineers journal. pattern analysis and machine intelligence, 1998,20 (2): method 113-125) is extracted the striation center in the pre-treatment sub-gratings, and calculate the coding at striation center according to the coding pattern, belong to any root striation thereby can discern the striation center.
4. repeat 2., 3., segment the identification of sub-gratings striations afterwards up to finishing all, thereby finish the identification of original high dencity grating striation.
Embodiment
Original grating adopts 128 striations as shown in Figure 3, and a display is measured.Obviously, because striation is too close, almost join together, identification brings very big difficulty to striation.Therefore we adopt pyramid sub-gratings projective method to throw two width of cloth sub-gratings, and are as Fig. 4, shown in Figure 5.Then sub-gratings 1 and sub-gratings 2 are encoded with the two fens compiling methods in space respectively.Fig. 6-11 is depicted as the pattern that sub-gratings 1 is carried out the two fens codings in space, and striation is respectively 32,16,8,4,2,1 at interval.
Employing history Dege (Ka Sideng. the history Dege, a kind of no inclined to one side detector of curvilinear structures, international Institute of Electrical and Electric Engineers journal. pattern analysis and machine intelligence, 1998,20 (2): method 113-125) is extracted the striation center of sub-gratings 1, according to six width of cloth coded images, determine six scale-of-two code value (Wang Zhao of striation central point with the method for Wang Zhao proposition, Tan Yushan. space encoding high speed three-dimensional profilometry----principle and method, laser and infrared, Vol.28, No.4,1998, pp.238-242).In like manner, six scale-of-two code values of sub-gratings 2 striation central points also can be determined.In order to distinguish the code value of sub-gratings 1 and sub-gratings 2 corresponding striations, the encoded radio of final original striation is introduced the 7th binary code, " 0 " expression sub-gratings 1, " 1 " expression sub-gratings 2.Thereby finish the coding of all striations.If grating subdivision is M=2
n(n is a positive integer) width of cloth sub-gratings is then discerned each sub-gratings with the n bit.
Claims (1)
1, a kind of grating projection and recognition methods, use digital projector as the raster pattern projector, digital projector is connected with a computing machine, produce the grating of arbitrary patterns by the computer control raster pattern projector, determine the density of the original grating that throws according to measuring needs, it is characterized in that
1.1, adopt " pyramid sub-gratings projective method " that raster density is reached the original grating that striation can not resolution degree to segment, establishing original grating has N root striation, N=2
n, n is a positive integer, the concrete steps of segmentation are:
1.1.1, every a striation original grating is subdivided into two width of cloth sub-gratings, every width of cloth sub-gratings contains N/2 root striation, the invariant position of every striation in original grating, segmentation back reach to differentiate as if sub-gratings and require promptly to stop segmentation, otherwise proceed next step segmentation;
1.1.2, to the segmentation after two width of cloth sub-gratings, continue to adopt the method for 1.1.1 to segment, original grating is subdivided into four width of cloth sub-gratings, every width of cloth sub-gratings contains N/4 root striation, the segmentation back requires promptly to stop segmentation if sub-gratings reaches to differentiate, otherwise proceed next step segmentation, requirement promptly stops segmentation if sub-gratings reaches resolution in the segmentation back, otherwise proceeds next step segmentation;
1.1.3, repeat the fine division step of 1.1.1, till the striation in the sub-gratings after segmentation can be differentiated identification;
1.2, to the segmentation after every width of cloth sub-gratings in striation discern;
Adopt the two fens compiling methods in space to carry out striation identification, concrete steps are:
1.2.1, determine the striation number in the sub-gratings;
If original grating striation number is N, the segmentation number is M, and then the striation number in the sub-gratings of segmentation back is: N
Sub=N/M;
1.2.2, projection encode grating pattern;
For any width of cloth sub-gratings, every N
Sub/ 2, N
Sub/ 4 ..., up to throw a width of cloth encode grating pattern every 1 striation, need projection 1+1og altogether
2(N
Sub) width of cloth coding pattern;
1.2.3, extract the striation center in the pre-treatment sub-gratings;
Adopt the method for history Dege to extract the striation center of working as in the pre-treatment sub-gratings, and calculate the coding at striation center, finish the coding of all striations in the pre-treatment sub-gratings according to the coding pattern;
1.2.4, repeating step 1.2.2 and 1.2.3, up to the identification of finishing all segmentation back sub-gratings striations, thereby finish the identification of original high dencity grating striation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100958914A CN100491906C (en) | 2005-09-07 | 2005-09-07 | Grating projection and recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100958914A CN100491906C (en) | 2005-09-07 | 2005-09-07 | Grating projection and recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1928491A CN1928491A (en) | 2007-03-14 |
| CN100491906C true CN100491906C (en) | 2009-05-27 |
Family
ID=37858561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100958914A Expired - Fee Related CN100491906C (en) | 2005-09-07 | 2005-09-07 | Grating projection and recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100491906C (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102798354B (en) * | 2012-08-28 | 2014-11-05 | 西北工业大学 | Binary stripe stack based sinusoidal grating generation method |
-
2005
- 2005-09-07 CN CNB2005100958914A patent/CN100491906C/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| 利用结构光进行三维测量的新方法. 虞启琏，张天虎.应用光学，第21卷第1期. 2000 |
| 利用结构光进行三维测量的新方法. 虞启琏，张天虎.应用光学，第21卷第1期. 2000 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1928491A (en) | 2007-03-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109887028B (en) | Unmanned vehicle auxiliary positioning method based on point cloud data registration | |
| CN101608906B (en) | Method for measuring optical three-dimensional contour of space coding | |
| CN109285198B (en) | Method for coding and identifying annular coding mark points | |
| CN101975558B (en) | Rapid three-dimensional measurement method based on color grating projection | |
| US4545067A (en) | Process for automatic image recognition | |
| CN103530889A (en) | Method for decoding coding mark points of measuring stick based on skeleton extraction | |
| CN111179187B (en) | Single image rain removing method based on cyclic generation countermeasure network | |
| CN111207692B (en) | Improved segmented step phase coding three-dimensional measurement method | |
| CN105260737A (en) | Automatic laser scanning data physical plane extraction method with multi-scale characteristics fused | |
| CN101504277A (en) | Method for acquiring object three-dimensional image by optical three-dimensional sensing | |
| CN103148876A (en) | Virtual incremental code channel construction method and decoding method for single-code-channel absolute linear-displacement grating ruler | |
| CN100491906C (en) | Grating projection and recognition method | |
| CN113192126A (en) | Distance measurement positioning method based on image circular object recognition | |
| CN104019834A (en) | Single code channel absolute position coding method and reading measuring system | |
| Zheng et al. | Joint coding strategy of the phase domain and intensity domain for absolute phase retrieval | |
| CN115060188A (en) | Frog point rail profile processing method and system | |
| CN116793247A (en) | Stripe projection profilometry stripe series correction method based on region statistics | |
| Pagès et al. | Implementation of a robust coded structured light technique for dynamic 3D measurements | |
| CN105224798B (en) | A kind of railway curve morpheme parameter identification method based on versed sine figure and angle figure | |
| CN101290217A (en) | Color coding structural light three-dimensional measurement method based on green stripe center | |
| CN104636708A (en) | Partial document image comparison method and system | |
| CN101833775A (en) | Encoding method of light image with color structure | |
| Mara et al. | Orientation of fragments of rotationally symmetrical 3d-shapes for archaeological documentation | |
| CN100449571C (en) | Threshold value dividing method based on single-pixel in three-dimensional scanning system | |
| EP1342992B1 (en) | Position transducer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090527 Termination date: 20120907 |