CN109376746A - A kind of image identification method and system - Google Patents
A kind of image identification method and system Download PDFInfo
- Publication number
- CN109376746A CN109376746A CN201811250479.9A CN201811250479A CN109376746A CN 109376746 A CN109376746 A CN 109376746A CN 201811250479 A CN201811250479 A CN 201811250479A CN 109376746 A CN109376746 A CN 109376746A
- Authority
- CN
- China
- Prior art keywords
- picture
- pictures
- sub
- identified
- association
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/752—Contour matching
Abstract
The present invention provides a kind of image identification method and system, comprising the following steps: S1, obtains picture to be identified；S2, picture to be identified is cut into several sub-pictures；S3, the characteristic information for extracting each sub-pictures；S4, the identification information in the characteristic information of each sub-pictures and identification library is compared, obtains the pre-selection association picture of picture to be identified, and each pre-selection association picture deposit is associated with picture group；There is association picture of the most picture of frequency as picture to be identified in S5, extraction association picture group；S6, the description information for extracting association picture simultaneously export recognition result.The identification work to picture to be identified can be completed using template similar with picture to be identified, to promote picture recognition success rate in identifying library when the not template picture completely the same with picture to be identified.The present invention is applied to picture process field.
Description
Technical field
The present invention relates to picture process field more particularly to a kind of image identification methods and system.
Background technique
Picture recognition refers to that pictorial stimulus acts on sense organ, people recognize it be a certain figure that experience is crossed mistake
Journey also makes picture re-recognize.In picture recognition, there should be the information for entering sense organ at that time, also to there is the information stored in memory.
The process being only compared by the information stored with current information, is just able to achieve and re-recognizes to picture.
Picture recognition technology is a key areas of artificial intelligence.In order to work out the simulation movable meter of mankind's picture recognition
Calculation machine program, there has been proposed different picture recognition models.Such as Template matching model.This model is thought, identifies some
Picture, it is necessary to have the memory pattern of this picture in past experience, be called template.If current stimulation can in brain
Template match, this picture is also just identified.Such as there are a letter A, if there is a A template in brain, alphabetical A's
Size, orientation, shape are all completely the same with this A template, and alphabetical A is just identified.Pattern-recognition in picture recognition
It (PatternRecognition), is a kind of from bulk information and data, in expertise and on the basis of be recognized,
Identification is automatically performed to shape, mode, curve, number, character format and figure using the method for computer and mathematical reasoning, is commented
The process of valence.Pattern-recognition includes two stages, i.e. study stage and implementation phase, the former is to carry out feature selecting to sample,
The rule of classification is found, the latter is that unknown sample collection is classified and identified according to classification rule, and final output identifies in library
The description information of template picture is as recognition result.The Template matching model of this pattern-recognition is simple and clear, is also easy to get
Practical application.But in the actual operation process, often have picture to be identified to identify library in template picture it is similar, but
It is again not quite identical, so as to cause picture recognition failure.
Summary of the invention
In view of the deficienciess of the prior art, the object of the present invention is to provide a kind of image identification methods.It can identify
In library not the template picture completely the same with picture to be identified when, using template similar with picture to be identified completion treat knowledge
The identification work of other picture, to promote picture recognition success rate.
The technical solution adopted by the present invention is that: a kind of image identification method, comprising the following steps:
S1, picture to be identified is obtained；
S2, picture to be identified is cut into several sub-pictures；
S3, the characteristic information for extracting each sub-pictures；
S4, the characteristic information of each sub-pictures and the identification information in identification library are compared, obtains the pre-selection of picture to be identified
It is associated with picture, and each pre-selection association picture deposit is associated with picture group；
There is association picture of the most picture of frequency as picture to be identified in S5, extraction association picture group；
S6, the description information for extracting association picture simultaneously export recognition result.
As a further improvement of the above technical scheme, in step S3, the characteristic information for extracting each sub-pictures includes:
It extracts the contour feature information of each sub-pictures and extracts the color characteristic information of each sub-pictures.
As a further improvement of the above technical scheme, the contour feature information for extracting each sub-pictures specifically includes:
S311, reading need each sub-pictures of contours extract；
S312, each sub-pictures are converted into LUV color space from rgb color space；
S313, it is split in color of the LUV color space to each sub-pictures；
S314, each sub-pictures are converted into grayscale image on the basis of picture segmentation；
S315, profile is obtained to the grayscale image progress edge detection of each sub-pictures；
S316, the form parameter for extracting each sub-pictures profile simultaneously quantitatively divide each sub-pictures according to the form parameter
Analysis.
As a further improvement of the above technical scheme, in step S313, using Pyramid technology algorithm to each sub-pictures
Color be split.
As a further improvement of the above technical scheme, the color characteristic information for extracting each sub-pictures includes:
Calculate the rgb value of mass-tone and the rgb value for calculating assertive colours in sub-pictures in each sub-pictures.
As a further improvement of the above technical scheme, the color characteristic information for extracting each sub-pictures specifically includes:
S321, reading need each sub-pictures of color extraction；
S322, the rgb value for obtaining each pixel in each sub-pictures；
S323, the average RGB value for calculating each pixel RGB values in sub-pictures, average RGB value is mass-tone in sub-pictures
Rgb value；
S324, difference between each pixel RGB values and average RGB value in sub-pictures is calculated, when difference maximum is corresponding
Rgb value is the rgb value of assertive colours in sub-pictures.
As a further improvement of the above technical scheme, in step S2, knowledge is treated using the method for scanning window or candidate window
Other picture is cut.
As a further improvement of the above technical scheme, in step S4, the acquisition condition of pre-selection association picture is: identification library
The identification information of middle template picture covers the characteristic information of any sub-pictures, which is that the pre-selection of picture to be identified is closed
Join picture.
As a further improvement of the above technical scheme, further include step S7, by after the completion of identification picture and its description
Information deposit identification library.
The present invention also provides a kind of picture recognition systems, the technical solution adopted is that:
A kind of picture recognition system, including memory and processor, the memory are stored with computer program, the place
Manage the step of realizing the above method when device executes the computer program.
Advantageous effects of the invention:
The present invention then extracts the contour feature of each sub-pictures by the way that picture to be identified is first cut into several sub-pictures,
Filter out in identification library again is used as pre-selection to be associated with picture with the template picture of the Patch-based match of each sub-pictures, and will be each pre-
Choosing association picture deposit association picture group, finishing screen, which is selected, there is the most template picture of frequency as wait know in associated diagram piece group
The association picture of other picture, and then export recognition result, can in identification library the not no mould completely the same with picture to be identified
When plate picture, the identification work to picture to be identified is completed using template similar with picture to be identified, to promote picture knowledge
Other success rate.
Detailed description of the invention
Fig. 1 is the flow diagram of image identification method in the present embodiment；
Fig. 2 is the structural schematic diagram for originally implementing middle picture recognition system.
Specific embodiment
A kind of image identification method as shown in Figure 1, comprising the following steps:
S1, obtain picture to be identified, in this implementations, can by camera or search engine input client or or its
His mode carries out picture input to be identified.
S2, picture to be identified is cut into several sub-pictures, using the method for scanning window or candidate window to picture to be identified
It is cut, such as the picture to be identified of a face is cut, then can obtain multiple local sons such as eye, mouth, nose, ear
Picture.
S3, the characteristic information for extracting each sub-pictures specially include extracting the contour feature information of each sub-pictures and extracting
The color characteristic information of each sub-pictures:
The contour feature information for extracting each sub-pictures specifically includes:
S311, reading need each sub-pictures of contours extract；
S312, each sub-pictures are converted into LUV color space from rgb color space；
S313, it is split in color of the LUV color space to each sub-pictures, is carried out here using Pyramid technology algorithm
Segmentation；
S314, each sub-pictures are converted into grayscale image on the basis of picture segmentation；
S315, profile is obtained to the grayscale image progress edge detection of each sub-pictures；
S316, the form parameter for extracting each sub-pictures profile simultaneously quantitatively divide each sub-pictures according to the form parameter
Analysis.
The color characteristic information for extracting each sub-pictures includes calculating the rgb value of mass-tone in each sub-pictures and calculating in sub-pictures
The rgb value of assertive colours, specifically:
S321, reading need each sub-pictures of color extraction；
S322, the rgb value for obtaining each pixel in each sub-pictures；
S323, the average RGB value for calculating each pixel RGB values in sub-pictures, average RGB value is mass-tone in sub-pictures
Rgb value；
S324, difference between each pixel RGB values and average RGB value in sub-pictures is calculated, when difference maximum is corresponding
Rgb value is the rgb value of assertive colours in sub-pictures.
S4, the characteristic information of each sub-pictures and the identification information in identification library are compared, obtains the pre-selection of picture to be identified
It is associated with picture, and each pre-selection association picture deposit is associated with picture group:
Here the identification information in identification library is the identification information of template picture, including contour feature, color characteristic
Deng when the identification information for having a template picture covers the characteristic information of any one sub-pictures, that is, covering any sub-pictures
Contour feature information or color characteristic information when, then by this template picture be classified as pre-selection association picture, deposit association picture
In group.
There is association picture of the most picture of frequency as picture to be identified in S5, extraction association picture group；
S6, the description information for extracting association picture simultaneously export recognition result；
S7, by after the completion of identification picture and its description information deposit identification library: by will identification complete picture also deposit
Enter to identify in library, to increase the template picture amount of storage in identification library.
The present invention then extracts the contour feature of each sub-pictures by the way that picture to be identified is first cut into several sub-pictures,
Filter out in identification library again is used as pre-selection to be associated with picture with the template picture of the Patch-based match of each sub-pictures, and will be each pre-
Choosing association picture deposit association picture group, finishing screen, which is selected, there is the most template picture of frequency as wait know in associated diagram piece group
The association picture of other picture, and then export recognition result, can in identification library the not no mould completely the same with picture to be identified
When plate picture, the identification work to picture to be identified is completed using template similar with picture to be identified, to promote picture knowledge
Other success rate.
The present embodiment additionally provides a kind of picture recognition system, the technical solution adopted is that:
A kind of picture recognition system as shown in Figure 2, including memory and processor, the memory are stored with computer
The step of program, the processor realizes the above method when executing the computer program, wherein processor includes that picture obtains
Module 1, contour feature extraction module 3, color feature extracted module 4, association picture identification module 5, is closed at picture cutting module 2
Join picture extraction module 6 and picture recognition module 7, picture uploading module 8, wherein picture cutting module 2, contour feature extract mould
Block 3, color feature extracted module 4, association picture identification module 5, association picture extraction module 6, picture uploading module 8 are integrated in
In the software module of CPU.
Picture obtains module 1, and for obtaining picture to be identified, in this implementations, picture acquisition module 1 can be camera
It can be search engine input client.
Picture cutting module 2 obtains module 1 with picture and is connected, for picture to be identified to be cut into several sub-pictures,
In by transferring the function of software realization picture cutting, cutting can be carried out using the method for scanning window or candidate window, such as right
The picture to be identified of one face carries out cutting, then can obtain multiple local sub-pictures such as eye, mouth, nose, ear.
Contour feature extraction module 3 is connected with picture cutting module 2, for extracting the contour feature of each sub-pictures, wherein
By the function of transferring the extraction of software realization contour feature.
Color feature extracted module 4 is connected with picture cutting module 2, for extracting the color characteristic of each sub-pictures, wherein
By the function of transferring software realization color feature extracted.
It is associated with picture identification module 5, is connected with contour feature extraction module 3, color feature extracted module 4, being used for will be each
Contour feature, the color characteristic of sub-pictures are compared with the identification information in identification library 9, and screen selecting formwork picture is made from identification library 9
It is associated with picture for the pre-selection of picture to be identified, and by each pre-selection association picture deposit association picture group, wherein by transferring software
Realize the function of association picture recognition, the identification information in identification library here is the identification information of template picture, including wheel
Wide feature, color characteristic etc., when the identification information for having a template picture covers the contour feature or face of any one sub-pictures
When color characteristic, then this template picture is classified as pre-selection association picture, deposit is associated in picture group.
It is associated with picture extraction module 6, is connected with picture identification module 5 is associated with, frequency occurs for extracting in association picture group
Association picture of the most picture of number as picture to be identified, wherein being associated with the function that picture extracts by transferring software realization.
Picture recognition module 7 is connected with picture extraction module 6 is associated with, and extracts the description information of association picture and exports knowledge
Other result.
Picture uploading module 8 obtains module 1 with picture, identifies that library 9, picture recognition module 7 are connected, for that will identify
Picture and its description information deposit identification library 9 after, to increase the template picture amount of storage in identification library 9.
Contour feature extraction module 3 includes:
Second picture reading submodule 31 is connected with picture cutting module 2, for reading each subgraph for needing contours extract
Piece；
Picture transform subblock 32 is connected with second picture reading submodule 31, for each sub-pictures are empty from rgb color
Between be converted into LUV color space, each sub-pictures are converted into LUV color space from rgb color space by transferring software；
Picture segmentation submodule 33 is connected with picture transform subblock 32, in LUV color space to each sub-pictures
Color is split, and is split by transferring software in color of the LUV color space to each sub-pictures；
Gray proces submodule 34 is connected with picture segmentation submodule 33, is used for each son on the basis of picture segmentation
Picture is converted into grayscale image, and each sub-pictures are converted into grayscale image on the basis of picture segmentation by transferring software；
Contours extract submodule 35 is connected with gray proces submodule 34, carries out side for the grayscale image to each sub-pictures
Edge detects to obtain profile, obtains profile to the grayscale image progress edge detection of each sub-pictures by transferring software.
Form parameter handles submodule 36, is connected with contours extract submodule 35, for extracting the shape of each sub-pictures profile
Shape parameter simultaneously carries out quantitative analysis to each sub-pictures according to form parameter.
Color feature extracted module 4 includes:
First picture reading submodule 41 is connected with picture cutting module 2, for reading each subgraph for needing color extraction
Piece；
Data acquisition submodule 42 is connected with the first picture reading submodule 41, for obtaining each picture in each sub-pictures
The rgb value of element, obtains the rgb value of each pixel in each sub-pictures by transferring software here；
First computational submodule 43, is connected with data acquisition submodule 42, for calculating the RGB of mass-tone in each sub-pictures
Value, the average RGB value of each pixel RGB values in each sub-pictures is calculated by transferring software, average RGB value is in sub-pictures
The rgb value of mass-tone；
Second computational submodule 44, is connected with data acquisition submodule 42, for calculating the RGB of assertive colours in each sub-pictures
Value calculates difference in each sub-pictures between each pixel RGB values and average RGB value by transferring software, when difference maximum
Corresponding rgb value is the rgb value of assertive colours in sub-pictures.
Contain the explanation of the preferred embodiment of the present invention above, this be for the technical characteristic that the present invention will be described in detail, and
Be not intended to for summary of the invention being limited in concrete form described in embodiment, according to the present invention content purport carry out other
Modifications and variations are also protected by this patent.The purport of the content of present invention is to be defined by the claims, rather than by embodiment
Specific descriptions are defined.
Claims (10)
1. a kind of image identification method, which comprises the following steps:
S1, picture to be identified is obtained；
S2, picture to be identified is cut into several sub-pictures；
S3, the characteristic information for extracting each sub-pictures；
S4, the characteristic information of each sub-pictures and the identification information in identification library are compared, obtains the pre-selection association of picture to be identified
Picture, and each pre-selection association picture deposit is associated with picture group；
There is association picture of the most picture of frequency as picture to be identified in S5, extraction association picture group；
S6, the description information for extracting association picture simultaneously export recognition result.
2. picture recognition system according to claim 1, which is characterized in that in step S3, the spy for extracting each sub-pictures
Reference ceases
It extracts the contour feature information of each sub-pictures and extracts the color characteristic information of each sub-pictures.
3. picture recognition system according to claim 2, which is characterized in that the contour feature information for extracting each sub-pictures
It specifically includes:
S311, reading need each sub-pictures of contours extract；
S312, each sub-pictures are converted into LUV color space from rgb color space；
S313, it is split in color of the LUV color space to each sub-pictures；
S314, each sub-pictures are converted into grayscale image on the basis of picture segmentation；
S315, profile is obtained to the grayscale image progress edge detection of each sub-pictures；
S316, the form parameter for extracting each sub-pictures profile simultaneously carry out quantitative analysis to each sub-pictures according to the form parameter.
4. picture recognition system according to claim 3, which is characterized in that in step S313, using Pyramid technology algorithm
The color of each sub-pictures is split.
5. picture recognition system according to claim 2, which is characterized in that the color characteristic information for extracting each sub-pictures
Include:
Calculate the rgb value of mass-tone and the rgb value for calculating assertive colours in sub-pictures in each sub-pictures.
6. picture recognition system according to claim 5, which is characterized in that the color characteristic information for extracting each sub-pictures
It specifically includes:
S321, reading need each sub-pictures of color extraction；
S322, the rgb value for obtaining each pixel in each sub-pictures；
S323, the average RGB value for calculating each pixel RGB values in sub-pictures, average RGB value is the RGB of mass-tone in sub-pictures
Value；
S324, calculate difference between each pixel RGB values and average RGB value in sub-pictures, when difference maximum corresponding rgb value
The rgb value of assertive colours as in sub-pictures.
7. according to claim 1 to any one of 6 picture recognition systems, which is characterized in that in step S2, using scanning window or
The method of candidate window cuts picture to be identified.
8. according to claim 1 to any one of 6 picture recognition systems, which is characterized in that in step S4, pre-selection association picture
Acquisition condition be: the identification information of template picture covers the characteristic informations of any sub-pictures in identification library, which is
Picture is associated with for the pre-selection of picture to be identified.
9. according to claim 1 to any one of 6 picture recognition systems, which is characterized in that further include step S7, will identify
Picture and its description information deposit identification library after.
10. a kind of picture recognition system, including memory and processor, the memory are stored with computer program, feature
It is, the step of processor realizes any one of claims 1 to 9 the method when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811250479.9A CN109376746A (en) | 2018-10-25 | 2018-10-25 | A kind of image identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811250479.9A CN109376746A (en) | 2018-10-25 | 2018-10-25 | A kind of image identification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109376746A true CN109376746A (en) | 2019-02-22 |
Family
ID=65401385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811250479.9A Pending CN109376746A (en) | 2018-10-25 | 2018-10-25 | A kind of image identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109376746A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523958A (en) * | 2020-03-10 | 2020-08-11 | 携程计算机技术（上海）有限公司 | Hotel fax return processing method, system, electronic equipment and medium |
| CN112288195A (en) * | 2020-11-24 | 2021-01-29 | 北京智源人工智能研究院 | Method and device for predicting central user behavior based on friend behavior and electronic equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196994A (en) * | 2007-12-26 | 2008-06-11 | 腾讯科技（深圳）有限公司 | Image content recognizing method and recognition system |
| CN102567952A (en) * | 2010-12-16 | 2012-07-11 | 阿里巴巴集团控股有限公司 | Image segmentation method and system |
| CN103186904A (en) * | 2011-12-31 | 2013-07-03 | 北京新媒传信科技有限公司 | Method and device for extracting picture contours |
| CN103605957A (en) * | 2013-11-05 | 2014-02-26 | 天脉聚源(北京)传媒科技有限公司 | Image identification method and device thereof |
| CN104899586A (en) * | 2014-03-03 | 2015-09-09 | 阿里巴巴集团控股有限公司 | Method for recognizing character contents included in image and device thereof |
| CN105741327A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Method and apparatus for extracting dominant color and assertive color of picture |
| CN107622250A (en) * | 2017-09-27 | 2018-01-23 | 深圳市得色科技有限公司 | 3D rendering recognition methods and its system based on machine learning |
| CN107751355A (en) * | 2017-09-05 | 2018-03-06 | 鲁东大学 | Materials flow chart predicts automatic control fruits and vegetables transportation resources as maturity |
| CN108197644A (en) * | 2017-12-27 | 2018-06-22 | 深圳市大熊动漫文化有限公司 | A kind of image-recognizing method and device |
| CN108416814A (en) * | 2018-02-08 | 2018-08-17 | 广州大学 | Quick positioning and recognition methods and the system on a kind of pineapple head |
-
2018
- 2018-10-25 CN CN201811250479.9A patent/CN109376746A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101196994A (en) * | 2007-12-26 | 2008-06-11 | 腾讯科技（深圳）有限公司 | Image content recognizing method and recognition system |
| CN102567952A (en) * | 2010-12-16 | 2012-07-11 | 阿里巴巴集团控股有限公司 | Image segmentation method and system |
| CN103186904A (en) * | 2011-12-31 | 2013-07-03 | 北京新媒传信科技有限公司 | Method and device for extracting picture contours |
| CN103605957A (en) * | 2013-11-05 | 2014-02-26 | 天脉聚源(北京)传媒科技有限公司 | Image identification method and device thereof |
| CN104899586A (en) * | 2014-03-03 | 2015-09-09 | 阿里巴巴集团控股有限公司 | Method for recognizing character contents included in image and device thereof |
| CN105741327A (en) * | 2014-12-10 | 2016-07-06 | 阿里巴巴集团控股有限公司 | Method and apparatus for extracting dominant color and assertive color of picture |
| CN107751355A (en) * | 2017-09-05 | 2018-03-06 | 鲁东大学 | Materials flow chart predicts automatic control fruits and vegetables transportation resources as maturity |
| CN107622250A (en) * | 2017-09-27 | 2018-01-23 | 深圳市得色科技有限公司 | 3D rendering recognition methods and its system based on machine learning |
| CN108197644A (en) * | 2017-12-27 | 2018-06-22 | 深圳市大熊动漫文化有限公司 | A kind of image-recognizing method and device |
| CN108416814A (en) * | 2018-02-08 | 2018-08-17 | 广州大学 | Quick positioning and recognition methods and the system on a kind of pineapple head |
Non-Patent Citations (1)
| Title |
|---|
| 王宇新: "基于特征分布的图像识别方法研究与应用", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523958A (en) * | 2020-03-10 | 2020-08-11 | 携程计算机技术（上海）有限公司 | Hotel fax return processing method, system, electronic equipment and medium |
| CN112288195A (en) * | 2020-11-24 | 2021-01-29 | 北京智源人工智能研究院 | Method and device for predicting central user behavior based on friend behavior and electronic equipment |
| CN112288195B (en) * | 2020-11-24 | 2021-08-31 | 北京智谱华章科技有限公司 | Method and device for predicting central user behavior based on friend behavior and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107958230B (en) | Facial expression recognition method and device | |
| CN110163069B (en) | Lane line detection method for driving assistance | |
| CN109002755B (en) | Age estimation model construction method and estimation method based on face image | |
| CN104021384B (en) | A kind of face identification method and device | |
| CN107292314A (en) | A kind of lepidopterous insects species automatic identification method based on CNN | |
| CN111080670A (en) | Image extraction method, device, equipment and storage medium | |
| CN107944398A (en) | Based on depth characteristic association list diagram image set face identification method, device and medium | |
| CN109583452A (en) | A kind of identification of individual method and system based on barefoot footprint | |
| CN112633221A (en) | Face direction detection method and related device | |
| CN109376746A (en) | A kind of image identification method and system | |
| Althabhawee et al. | Fingerprint recognition based on collected images using deep learning technology | |
| CN104463091A (en) | Face image recognition method based on LGBP feature subvectors of image | |
| CN111612083B (en) | Finger vein recognition method, device and equipment | |
| CN110363762B (en) | Cell detection method, cell detection device, intelligent microscope system and readable storage medium | |
| CN112101293A (en) | Facial expression recognition method, device, equipment and storage medium | |
| Vezzetti et al. | Application of geometry to rgb images for facial landmark localisation-a preliminary approach | |
| Karungaru et al. | Face recognition in colour images using neural networks and genetic algorithms | |
| CN109961010A (en) | A kind of gesture identification method based on intelligent robot | |
| Kalangi et al. | Deployment of Haar Cascade algorithm to detect real-time faces | |
| Yani et al. | A robust damaged fingerprint identification algorithm based on deep learning | |
| CN115705748A (en) | Facial feature recognition system | |
| CN111582014A (en) | Container identification method, device and computer readable storage medium | |
| CN112733670A (en) | Fingerprint feature extraction method and device, electronic equipment and storage medium | |
| Christabel et al. | Facial feature extraction based on local color and texture for face recognition using neural network | |
| Kocevar et al. | Low-quality fingerprint image enhancement on the basis of oriented diffusion and ridge compensation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |