CN103996015A - Method and apparatus for 3D image identification - Google Patents
Method and apparatus for 3D image identification Download PDFInfo
- Publication number
- CN103996015A CN103996015A CN201310443404.3A CN201310443404A CN103996015A CN 103996015 A CN103996015 A CN 103996015A CN 201310443404 A CN201310443404 A CN 201310443404A CN 103996015 A CN103996015 A CN 103996015A
- Authority
- CN
- China
- Prior art keywords
- image
- rendering
- angle point
- corner detection
- gray level
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
Abstract
The invention provides a method and apparatus for 3D image identification. The method comprises the following steps that: a, an image is inputted; b, the image is processed into a left gray level image and a right gray level image; c, a corner detection algorithm is used for carrying out processing on the left gray level image and the right gray level image to respectively solve corners; d, a corner matching algorithm is used for carrying out matching on the corners of the two images; e, an elimination algorithm is used for eliminating a wrong matched corner; and f, whether the number of pairs of matched corners of the two images is larger than or equal to N is determined; if so, the image is a 3D image; and if not, the image is a non-3D image. According to the invention, with the method and the apparatus, the 3D image and the non-3D image can be distinguished automatically without the need for the user to carry out eye distinguishing, thereby saving the time and energy of the user.
Description
Technical field
The present invention relates to image and process and field of image recognition, particularly relate to a kind of to 3D rendering knowledge method for distinguishing and device.
Background technology
Existing algorithm does not provide knows method for distinguishing to 3D rendering and non-3D rendering, and it is 3D rendering that people need to distinguish with visual inspection, and this method takes time and effort.
For solving the problems of the technologies described above, the method and the device that the invention provides a kind of full-automation are distinguished 3D rendering and non-3D rendering, without going careful resolution by people's naked eyes again.
Summary of the invention
The invention provides a kind of to 3D rendering knowledge method for distinguishing and device, by image being treated to left and right gray level image, the left and right gray level image after conversion being carried out to the rejecting of Corner Detection, corners Matching and wrong angle point, the final correct coupling angle point logarithm that obtains some, then the coupling angle point logarithm minimum corners Matching logarithm N corresponding with 3D rendering compared, automatic distinguishing 3D rendering and non-3D rendering, with the naked eye judge again without user, can save user's time and muscle power.
For solving the problems of the technologies described above, first technical scheme provided by the invention is to provide a kind of to 3D rendering knowledge method for distinguishing, comprises flow process:
A. input picture;
B. image is treated to the gray level image of left and right;
C. by Corner Detection Algorithm, left and right image is asked respectively to angle point;
D. with corner correspondence, the angle point of two width images is mated;
E. reject wrong coupling angle point with rejecting algorithm;
F. judge logarithm that the angle point of two width images matches whether >=N, if so, described image is 3D rendering, and if not, described image is non-3D rendering;
Wherein, N is that image is the 3D rendering minimum value of the coupling angle point logarithm of desired left and right image constantly.
Wherein, described Corner Detection Algorithm comprises Harris Corner Detection Algorithm.
Wherein, described rejecting algorithm comprises light beam method of adjustment.
According to the method described in above-mentioned any one, when the pixel of described image is 480*640, corresponding N=50.
For solving the problems of the technologies described above, second technical scheme provided by the invention is to provide a kind of device to 3D rendering identification, comprising:
Image conversion module, for being treated to the image of input the gray level image of left and right;
Corner Detection module, is electrically connected with described image conversion module, for utilizing Corner Detection Algorithm to ask respectively angle point to left and right image;
Corners Matching module, is electrically connected with described Corner Detection module, for utilizing corner correspondence to mate the angle point of two width images, and utilizes rejecting algorithm to reject wrong coupling angle point;
Judge module, is electrically connected with described corners Matching module, and for judging logarithm that the angle point of two width images matches whether >=N, if so, described image is 3D rendering, and if not, described image is non-3D rendering, and exports described judged result;
Wherein, N is the minimum value of image coupling angle point logarithm of desired left and right image while being 3D rendering.Wherein, described Corner Detection Algorithm comprises Harris Corner Detection Algorithm.Wherein, described rejecting algorithm comprises light beam method of adjustment.
The description of the device according to above-mentioned any one to 3D rendering identification, when the pixel of described image is 480*640, corresponding N=50.
The invention has the beneficial effects as follows: by the imagery exploitation transfer algorithm of input is converted to left and right gray level image, then carry out the rejecting of Corner Detection, corners Matching and wrong angle point, the angle point logarithm of the final correct coupling that obtains some, then the coupling angle point logarithm minimum corners Matching logarithm N corresponding with 3D rendering compared, automatic distinguishing 3D rendering and non-3D rendering, with the naked eye judge again without user, can save user's time and muscle power and test.
Brief description of the drawings
Fig. 1 is the present invention knows an embodiment of method for distinguishing schematic flow sheet to 3D rendering.
Fig. 2 is the structural representation of an embodiment of the device of the present invention to 3D rendering identification.
Embodiment
Refer to Fig. 1, Fig. 1 is the present invention knows an embodiment of method for distinguishing schematic flow sheet to 3D rendering.As shown in Figure 1, the present embodiment comprises flow process:
A. input picture;
B. image is treated to the gray level image of left and right,
Automatically the image of input is converted to gray level image by translation arithmetic, and the image after conversion is split into the picture format of left and right;
C. by Corner Detection Algorithm, left and right gray level image is asked respectively to angle point,
In an embodiment of the present invention, preferred Corner Detection Algorithm comprises Harris Corner Detection Algorithm, but also can use other can reach the Corner Detection Algorithm of same object, and the present invention is not restricted this;
D. with corner correspondence, the angle point of two width images is mated,
In the time utilizing Corner Detection Algorithm to detect the current location of angle point, program is called corner correspondence again the angle point of two width images is mated;
E. reject wrong coupling angle point with rejecting algorithm,
Because some angle point does not match; utilize rejecting algorithm can reject wrong coupling angle point; only leave the correct coupling angle point of some; concrete rejecting algorithm comprises light beam method of adjustment; certainly in concrete application, also can adopt other the rejecting algorithms that can realize same object to reach object of the present invention, the present invention does not also limit this;
F. judge logarithm that the angle point of two width images matches is whether >=N, if so, described image is 3D rendering, and if not, described image is non-3D rendering,
For the definite image of a width pixel, if 3D rendering, the coupling angle point logarithm of this image must be more than or equal to a constant being associated with image pixel, in embodiments of the invention, represent this constant by parameter N, N is also the minimum value of image corresponding images match angle point logarithm while being 3D rendering.For example, if the pixel 480*640 of image time, corresponding N=50, is more than or equal to 50 at the match point logarithm of the two width images that are divided into, corresponding image is 3D rendering, in this process step, utilize this constant N to judge whether the image of input is 3D rendering, particularly: judge logarithm that the angle point of two width images matches whether >=N, if, described image is 3D rendering, and if not, image is non-3D rendering.
Refer to Fig. 2, Fig. 2 is the schematic diagram structure of an embodiment of the device of the present invention to 3D rendering identification.As shown in Figure 2, the device 20 to 3D rendering identification of the present embodiment comprises image conversion module 21, Corner Detection module 22, corners Matching module 23 and judge module 24.Wherein, image conversion module 21 is electrically connected with Corner Detection module 22, Corner Detection module 22 is electrically connected with corners Matching module 23, corners Matching module 23 is electrically connected with judge module 24, and data flow to Corner Detection module 22, corners Matching module 23 and judge module 24 successively from image conversion module 21.
In the present embodiment, image conversion module 21 receives after the image of input, image is converted to the gray level image of left and right by transfer algorithm.Then, gray level image is transferred to Corner Detection module 22, and Corner Detection module 22 utilizes Corner Detection Algorithm to ask respectively angle point to obtain the position coordinates of angle point to left and right gray level image.Corner Detection module 22 utilizes corner correspondence to mate the angle point of two width images, and utilizes rejecting algorithm to reject wrong coupling angle point, and then obtains the correct coupling angle point logarithm of some.Judge module 24 judges logarithm that the angle point of two width images matches whether >=N, and if so, described image is 3D rendering, if not, described image is non-3D rendering, and exports described judged result, wherein, N is the minimum value of image coupling angle point logarithm of desired left and right image while being 3D rendering.
In a preferred embodiment of the present invention, Corner Detection Algorithm comprises Harris Corner Detection Algorithm.In a preferred embodiment of the present invention, reject algorithm and comprise light beam method of adjustment.Certainly, in other embodiments of the invention, also can realize the Corner Detection Algorithm of the object of the invention and reject algorithm with other, the present invention limit this.
In a preferred embodiment of the device to 3D rendering identification of the present invention, if when the image pixel that image conversion module 21 is received is 480*640, corresponding N=50, therefore be 480*640 image for pixel, when the angle point logarithm matching of left and right gray level image is more than or equal to 50, image is just 3D rendering, otherwise is non-3D rendering, and the device to 3D rendering identification of the present invention judges that taking this whether image is as 3D rendering.
By the way, provided by the invention 3D rendering is known to method for distinguishing and device can identify 3D rendering and non-3D rendering accurately, without user's visual inspection and then judge that whether image is 3D rendering, can save manpower and time.
The foregoing is only embodiments of the invention; not thereby limit the scope of the claims of the present invention; every equivalent structure or conversion of equivalent flow process that utilizes instructions of the present invention and accompanying drawing content to do; or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.
Claims (8)
1. 3D rendering is known to a method for distinguishing, it is characterized in that, described method comprises flow process:
A. input picture;
B. image is treated to the gray level image of left and right;
C. by Corner Detection Algorithm, left and right gray level image is asked respectively to angle point;
D. with corner correspondence, the angle point of two width images is mated;
E. reject wrong coupling angle point with rejecting algorithm;
F. judge logarithm that the angle point of two width images matches whether >=N, if so, described image is 3D rendering, and if not, described image is non-3D rendering;
Wherein, N is the minimum value of image coupling angle point logarithm of desired left and right image while being 3D rendering.
2. method according to claim 1, is characterized in that, described Corner Detection Algorithm comprises Harris Corner Detection Algorithm.
3. method according to claim 1, is characterized in that, described rejecting algorithm comprises light beam method of adjustment.
4. according to the method described in claim 1-3 any one, it is characterized in that, when the pixel of described image is 480*640, corresponding N=50.
5. the device to 3D rendering identification, is characterized in that, described device comprises:
Image conversion module, for being treated to the image of input the gray level image of left and right;
Corner Detection module, is electrically connected with described image conversion module, for utilizing Corner Detection Algorithm to ask respectively angle point to left and right gray level image;
Corners Matching module, is electrically connected with described Corner Detection module, for utilizing corner correspondence to mate the angle point of two width images, and utilizes rejecting algorithm to reject wrong coupling angle point;
Judge module, is electrically connected with described corners Matching module, and for judging logarithm that the angle point of two width images matches whether >=N, if so, described image is 3D rendering, and if not, described image is non-3D rendering, and exports described judged result;
Wherein, N is the minimum value of image coupling angle point logarithm of desired left and right image while being 3D rendering.
6. device according to claim 5, is characterized in that, described Corner Detection Algorithm comprises Harris Corner Detection Algorithm.
7. device according to claim 5, is characterized in that, described rejecting algorithm comprises light beam method of adjustment.
8. according to the device described in claim 5-7 any one, it is characterized in that, when the pixel of described image is 480*640, corresponding N=50.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310443404.3A CN103996015B (en) | 2013-09-26 | 2013-09-26 | A kind of method and device to 3D rendering identification |
| PCT/CN2014/082899 WO2015043301A1 (en) | 2013-09-26 | 2014-07-24 | Method and apparatus for recognizing 3d image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310443404.3A CN103996015B (en) | 2013-09-26 | 2013-09-26 | A kind of method and device to 3D rendering identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103996015A true CN103996015A (en) | 2014-08-20 |
| CN103996015B CN103996015B (en) | 2016-09-07 |
Family
ID=51310177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310443404.3A Expired - Fee Related CN103996015B (en) | 2013-09-26 | 2013-09-26 | A kind of method and device to 3D rendering identification |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN103996015B (en) |
| WO (1) | WO2015043301A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023171B (en) * | 2016-05-12 | 2019-05-14 | 惠州学院 | A kind of image angular-point detection method based on turning radius |
| WO2018187939A1 (en) * | 2017-04-11 | 2018-10-18 | 深圳市柔宇科技有限公司 | Method for identifying three-dimensional image, and terminal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625768A (en) * | 2009-07-23 | 2010-01-13 | 东南大学 | Three-dimensional human face reconstruction method based on stereoscopic vision |
| CN102074015A (en) * | 2011-02-24 | 2011-05-25 | 哈尔滨工业大学 | Two-dimensional image sequence based three-dimensional reconstruction method of target |
| CN103177468A (en) * | 2013-03-29 | 2013-06-26 | 渤海大学 | Three-dimensional motion object augmented reality registration method based on no marks |
| CN103247053A (en) * | 2013-05-16 | 2013-08-14 | 大连理工大学 | Accurate part positioning method based on binocular microscopy stereo vision |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2510499A1 (en) * | 2009-12-09 | 2012-10-17 | Thomson Licensing | Method for distinguishing a 3d image from a 2d image and for identifying the presence of a 3d image format by feature correspondence determination |
| CN101799939A (en) * | 2010-04-02 | 2010-08-11 | 天津大学 | Rapid and self-adaptive generation algorithm of intermediate viewpoint based on left and right viewpoint images |
| CN102395037B (en) * | 2011-06-30 | 2014-11-05 | 深圳超多维光电子有限公司 | Format recognition method and device |
| CN103051913A (en) * | 2013-01-05 | 2013-04-17 | 北京暴风科技股份有限公司 | Automatic 3D (three-dimensional) film source identification method |
-
2013
- 2013-09-26 CN CN201310443404.3A patent/CN103996015B/en not_active Expired - Fee Related
-
2014
- 2014-07-24 WO PCT/CN2014/082899 patent/WO2015043301A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101625768A (en) * | 2009-07-23 | 2010-01-13 | 东南大学 | Three-dimensional human face reconstruction method based on stereoscopic vision |
| CN102074015A (en) * | 2011-02-24 | 2011-05-25 | 哈尔滨工业大学 | Two-dimensional image sequence based three-dimensional reconstruction method of target |
| CN103177468A (en) * | 2013-03-29 | 2013-06-26 | 渤海大学 | Three-dimensional motion object augmented reality registration method based on no marks |
| CN103247053A (en) * | 2013-05-16 | 2013-08-14 | 大连理工大学 | Accurate part positioning method based on binocular microscopy stereo vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103996015B (en) | 2016-09-07 |
| WO2015043301A1 (en) | 2015-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Elrefaei et al. | Automatic electricity meter reading based on image processing | |
| CN105913093A (en) | Template matching method for character recognizing and processing | |
| CN102779274B (en) | Intelligent television face recognition method based on binocular camera | |
| CN102938077A (en) | Online AOI (Automatic Optical Inspection) image retrieval method based on double-threshold binaryzation | |
| CN103400124A (en) | Method and device for identifying test paper | |
| US20140050387A1 (en) | System and Method for Machine Vision Inspection | |
| CN103425984B (en) | A kind of detect the method and apparatus of regular polygonal seal in bill | |
| CN106033544A (en) | Test content area extraction method based on template matching | |
| CN105701764B (en) | A kind of joining method for medical image | |
| CN103914833A (en) | Method and system for automatically detecting whether bill is incomplete or not | |
| CN111985465A (en) | Text recognition method, device, equipment and storage medium | |
| CN112784853B (en) | Terminal connection state detection method and device | |
| CN105760869B (en) | Examination paper marking method based on image gray level recognition and processing | |
| CN104281850B (en) | character area identification method and device | |
| CN103150725B (en) | Based on SUSAN edge detection method and the system of non-local mean | |
| CN112991374A (en) | Canny algorithm-based edge enhancement method, device, equipment and storage medium | |
| CN106485639A (en) | The method and apparatus differentiating forged certificate picture | |
| CN103996015A (en) | Method and apparatus for 3D image identification | |
| Lay et al. | Rectification of QR-code images using the parametric cylindrical surface model | |
| CN105791815B (en) | A kind of TV line automatic judging methods | |
| CN111830051A (en) | Transformer substation equipment oil leakage detection method and detection system based on deep learning | |
| CN114418951A (en) | Pad detection method and device, computer equipment and storage medium | |
| JP2008084109A (en) | Eye opening/closing determination device and eye opening/closing determination method | |
| CN103837098A (en) | Screen test device and method | |
| CN106446902B (en) | Non-legible image-recognizing method and device |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160907 Termination date: 20200926 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |