CN104376310A - Face recognition method and device - Google Patents
Face recognition method and device Download PDFInfo
- Publication number
- CN104376310A CN104376310A CN201410707594.XA CN201410707594A CN104376310A CN 104376310 A CN104376310 A CN 104376310A CN 201410707594 A CN201410707594 A CN 201410707594A CN 104376310 A CN104376310 A CN 104376310A
- Authority
- CN
- China
- Prior art keywords
- facial image
- elementary contour
- described facial
- processed
- gray
- 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
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a face recognition method and device. The face recognition method includes acquiring a face image, processing the face image through a contour model, determining a first basic contour of the face image, and processing the first basic contour through a first skin color model to acquire a result of the face image recognition. By the use of the face recognition method integrating the skin color model and the geometrical characteristics, a covering on the face is accurately recognized, and accordingly the face is accurately recognized.
Description
Technical field
The present invention relates to technical field of image processing, particularly relate to a kind of face identification method and device thereof.
Background technology
Recognition of face is a kind of biological identification technology carrying out identification based on the face feature information of people.
In prior art, first by camera collection image, then, characteristics extraction is carried out to this image, wherein, those eigenwerts are proper vectors of reflection human face shape and geometric relationship, then, the eigenwert extracted is mated with the proper vector in template, if the match is successful, then identify successfully, and obtain face recognition result.
But, under glasses, hair or head jewelry block the scene of face, according to the method for prior art, then cannot find the proper vector of mating with image feature value in a template, thus recognition of face cannot be carried out accurately.
Summary of the invention
The invention provides a kind of face identification method and device thereof, the problem cannot carrying out recognition of face accurately can be solved.
First aspect, the embodiment of the present invention provides a kind of face identification method, comprising:
Obtain facial image;
Pre-service is carried out to described facial image, obtains pre-processed results; Described pre-service at least comprises: processed described facial image by skeleton pattern, determines the first elementary contour of described facial image; Described pre-processed results at least comprises described first elementary contour;
By the first complexion model, described pre-processed results is processed, obtain the recognition result of described facial image.
Optionally, describedly by skeleton pattern, described facial image to be processed, before determining the first elementary contour of described facial image, pre-service is carried out to facial image and also comprises:
Before determining the first elementary contour of described facial image, by the second complexion model, described facial image is processed, determine the second elementary contour of described facial image; Described pre-processed results also comprises described second elementary contour.
Optionally, describedly pre-service carried out to facial image also comprise:
Before the first elementary contour determining described facial image or the second elementary contour, pre-service gray proces is carried out to described facial image, obtains the pre-processed results of described facial image; Described gray proces comprises one of following: to described facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
Optionally, describedly by skeleton pattern, described facial image to be processed, determines the first elementary contour of described facial image, comprising:
Adopt sobel operator to process described facial image, determine the first elementary contour of described facial image.
Optionally, describedly by the first complexion model, described pre-processed results to be processed, obtains the recognition result of described facial image, comprising:
Adopt AdaBoost algorithm to process described pre-processed results, obtain the recognition result of described facial image.Second aspect, the embodiment of the present invention provides a kind of face identification device, comprising:
Acquisition module, for obtaining facial image;
Profile module, for carrying out pre-service to described facial image, obtains pre-processed results; Described pre-service at least comprises: processed described facial image by skeleton pattern, determines the first elementary contour of described facial image; Described pre-processed results at least comprises the first elementary contour that described first elementary contour determines described facial image;
First colour of skin module, for processing described pre-processed results first elementary contour, obtains the recognition result of described facial image.
Optionally, also comprise: the second colour of skin module;
Described second colour of skin module, for before determining the first elementary contour of described facial image, processes described facial image, determines the second elementary contour of described facial image; Described pre-processed results also comprises described second elementary contour
Described first colour of skin module, specifically for processing described first elementary contour and described second elementary contour, obtains the recognition result of described facial image.
Optionally, also comprise: gray scale pretreatment module;
Described pre-gradation processing module, before the first elementary contour of determining described facial image or the second elementary contour, carries out pre-gray proces to described facial image, obtains the pre-processed results of described facial image; Described gray proces comprises one of following: to described facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
Described first colour of skin module, specifically for processing described pre-processed results, described first elementary contour and described second elementary contour, obtains the recognition result of described facial image.
Optionally, described profile module, specifically for adopting sobel operator to process described facial image, determines the first elementary contour of described facial image.
Optionally, described first colour of skin module, enters pre-processed results specifically for adopting AdaBoost algorithm to described first elementary contour and processes, obtain the recognition result of described facial image.
The face identification method that the embodiment of the present invention provides and device thereof, by obtaining facial image, then, by skeleton pattern, facial image is processed, determine the first elementary contour of facial image, then, by the first complexion model, the first elementary contour is processed, obtain the recognition result of facial image.Achieve the face identification method combined based on complexion model and geometric properties, thus the overcover on face can be identified accurately, and then carry out recognition of face accurately.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of the present inventor's face recognition method one embodiment;
Fig. 2 is the schematic flow sheet of another embodiment of the present inventor's face recognition method;
Fig. 3 is the structural representation of face identification device one embodiment of the present invention;
Fig. 4 is the structural representation of another embodiment of face identification device of the present invention.
Embodiment
For making the object of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
When the face identification method that the embodiment of the present invention provides can be applied to and carry out recognition of face by face identification device.The face identification method that the present embodiment provides specifically can be performed by face identification device, this face identification device can be integrated in gate control system, body verification system, regional monitoring system or attendance checking system, and this face identification device can adopt the mode of software and/or hardware to realize.Below the face identification method that the present embodiment provides is described in detail.
Fig. 1 is the schematic flow sheet of the present inventor's face recognition method one embodiment, and as shown in Figure 1, the executive agent of the present embodiment is face identification device.The method comprises the steps:
Step 101, acquisition facial image.
For example, facial image can be obtained by camera.
In the present embodiment, facial image can be two-dimensional image information, such as, and the information such as gray matrix.
Step 102, pre-service is carried out to facial image, obtain pre-processed results.
In the present embodiment, pre-service at least comprises: processed facial image by skeleton pattern, determines the first elementary contour of facial image, and that is, pre-processed results at least comprises the first elementary contour.Preferably, in the present embodiment, sobel operator can be adopted to process facial image, determine the first elementary contour of facial image.
Optionally, the present embodiment by the right boundary of the gray-scale value determination face of facial image and up-and-down boundary, further, then can carry out this process by human face three-dimensional model to facial image again, thus determines the first elementary contour of facial image.
It should be noted that, the present embodiment repeats the concrete methods of realizing adopting skeleton pattern to carry out face image processing no longer one by one, those skilled in the art can adopt any one method in existing skeleton pattern to process facial image, thus obtain the first elementary contour of facial image, namely tentatively determine the Position Approximate of face and face.
Step 103, by the first complexion model, pre-processed results to be processed, obtain the recognition result of facial image.
Preferably, in the present embodiment, can in YCbCr model, the pre-processed results by facial image converts YCbCr coding to, and adopts AdaBoost algorithm to process the pre-processed results converting YCbCr coding to, obtains the recognition result of facial image.
Optionally, the present embodiment can in YUV model, and the pre-processed results by facial image converts YUV coding to, adopt vertical gray-level projection or level and smooth rear vertical gray-level projection, determine the exact position of face, and set up two-dimensional histogram, thus obtain the recognition result of facial image.
It should be noted that, the present embodiment repeats the concrete methods of realizing adopting complexion model to carry out face image processing no longer one by one, those skilled in the art can adopt any one method in existing complexion model to process pre-processed results, thus obtain the recognition result of facial image.
In the present embodiment, by obtaining facial image, then, pre-service being carried out to described facial image, obtains pre-processed results; Described pre-service at least comprises: processed facial image by skeleton pattern, determine the first elementary contour of facial image, wherein, pre-processed results at least comprises described first elementary contour, then, by the first complexion model, pre-processed results is processed, obtain the recognition result of facial image.Achieve the face identification method combined based on complexion model and geometric properties, thus the overcover on face can be identified accurately, and then carry out recognition of face accurately.
Fig. 2 is the schematic flow sheet of another embodiment of the present inventor's face recognition method, and as shown in Figure 2, the executive agent of the present embodiment is face identification device.The method comprises the steps:
Step 201, acquisition facial image.
The implementation of the present embodiment and step 101 similar, then this repeats no longer one by one.
Step 202, by the second complexion model, facial image to be processed, determine the second elementary contour of facial image.
For example, in the present embodiment, by YCbCr model, facial image is processed, to determine the second elementary contour of facial image.
Step 203, by skeleton pattern, facial image to be processed, determine the first elementary contour of facial image.
Preferably, in the present embodiment, sobel operator can be adopted to process facial image, determine the first elementary contour of facial image.
The present embodiment repeats the concrete methods of realizing adopting skeleton pattern to carry out face image processing no longer one by one, those skilled in the art can adopt any one method in existing skeleton pattern to process facial image, thus obtain the first elementary contour of facial image, namely tentatively determine the Position Approximate of face and face.
It should be noted that, in the method that the present embodiment provides, the realization of step 202 and step 203 in no particular order.
Step 204, by the first complexion model, the first elementary contour and the second elementary contour to be processed, obtain the recognition result of facial image.
Preferably, in the present embodiment, in YCbCr model, namely respectively the first elementary contour of facial image and the second elementary contour can be converted to YCbCr coding, and adopt AdaBoost algorithm to process the first elementary contour, obtain the recognition result of facial image.
It should be noted that, the present embodiment repeats the concrete methods of realizing adopting complexion model to carry out face image processing no longer one by one, those skilled in the art can adopt any one method in existing complexion model to process the first elementary contour and the second elementary contour, thus obtain the recognition result of facial image.
On the basis of above-described embodiment, namely before carrying out complexion model and geometric properties process, first complexion model process is carried out, further, carry out the recognition of face combined based on complexion model and geometric properties again, the overcover on face can be identified more accurately, add the robustness of recognition of face.
On the basis of above-described embodiment, before step 102 or step 203, before the first elementary contour namely determining facial image or the second elementary contour, can also comprise:
Gray proces is carried out to facial image.
Gray proces in the present embodiment can comprise one of following: to facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
In the present embodiment, gray-scale map conversion or gray-scale map histogram equalization can be carried out to the facial image obtained.Further, vertical gray-level projection, level and smooth rear vertical gray-level projection process can also be carried out.Thus acquisition is more easily processed, the facial image that degree of accuracy is higher.
Further, on the basis of above-described embodiment, by the first complexion model, the first elementary contour is processed, obtains the recognition result of facial image, can comprise:
By the first complexion model, pre-processed results, the first elementary contour and the second elementary contour are processed, obtain the recognition result of facial image.
In the present embodiment, by carrying out pre-service to the facial image obtained, the overcover on face can be identified more accurately, and add the robustness of recognition of face.
Fig. 3 is the structural representation of face identification device one embodiment of the present invention, and as shown in Figure 3, this face identification device, comprising: acquisition module 31, profile module 32 and the first colour of skin module 33, wherein,
Acquisition module 31, for obtaining facial image;
Profile module 32, for carrying out pre-service to facial image, obtains pre-processed results; Described pre-service at least comprises: processed described facial image by skeleton pattern, determines the first elementary contour of described facial image; Described pre-processed results at least comprises described first takes turns substantially;
First colour of skin module 33, for processing described pre-processed results, obtains the recognition result of described facial image.
In the present embodiment, by obtaining facial image, then, by skeleton pattern, facial image is processed, determine the first elementary contour of facial image, then, by the first complexion model, the first elementary contour is processed, obtain the recognition result of facial image.Achieve the face identification method combined based on complexion model and geometric properties, thus the overcover on face can be identified accurately, and then carry out recognition of face accurately.
Fig. 4 is the structural representation of another embodiment of face identification device of the present invention, and as shown in Figure 4, this face identification device, can also comprise: the second colour of skin module 34;
Described second colour of skin module 34, for before determining the first elementary contour of described facial image, processes described facial image, determines the second elementary contour of described facial image; Described pre-processed results also comprises described second elementary contour.
Further, on the basis of upper embodiment, this face identification device can also comprise: gradation processing module 35; Wherein,
Described gradation processing module 35, before the first elementary contour of determining described facial image or the second elementary contour, carries out gray proces to described facial image; Described gray proces comprises one of following: to described facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
Preferably, described profile module 32, specifically for adopting sobel operator to process described facial image, determines the first elementary contour of described facial image.
First colour of skin module 33, specifically for adopting AdaBoost algorithm to process described pre-processed results, obtains the recognition result of described facial image.
In the present embodiment, first pre-service is carried out to the facial image obtained, and by the recognition of face that complexion model and geometric properties combine, the overcover on face can be identified more accurately, and add the robustness of recognition of face.
One of ordinary skill in the art will appreciate that: all or part of step realizing above-mentioned each embodiment of the method can have been come by the hardware that programmed instruction is relevant.Aforesaid program can be stored in a computer read/write memory medium.This program, when performing, performs the step comprising above-mentioned each embodiment of the method; And aforesaid storage medium comprises: ROM, RAM, magnetic disc or CD etc. various can be program code stored medium.
Last it is noted that above each embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to foregoing embodiments to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein some or all of technical characteristic; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.
Claims (10)
1. a face identification method, is characterized in that, comprising:
Obtain facial image;
Pre-service is carried out to described facial image, obtains pre-processed results; Described pre-service at least comprises: processed described facial image by skeleton pattern, determines the first elementary contour of described facial image; Described pre-processed results at least comprises described first elementary contour;
By the first complexion model, described pre-processed results is processed, obtain the recognition result of described facial image.
2. by face identification method according to claim 1, it is characterized in that, described pre-service carried out to facial image, also comprise:
Before determining the first elementary contour of described facial image, by the second complexion model, described facial image is processed, determine the second elementary contour of described facial image; Described pre-processed results also comprises described second elementary contour.
3. by face identification method according to claim 2, it is characterized in that, described pre-service carried out to facial image, also comprise:
Before the first elementary contour determining described facial image or the second elementary contour, gray proces is carried out to described facial image; Described gray proces comprises one of following: to described facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
4. by face identification method described in any one of claim 1-3, it is characterized in that, describedly by skeleton pattern, described facial image to be processed, determine the first elementary contour of described facial image, comprising:
Adopt sobel operator to process described facial image, determine the first elementary contour of described facial image.
5. by face identification method according to claim 4, it is characterized in that, describedly by the first complexion model, described pre-processed results to be processed, obtain the recognition result of described facial image, comprising:
Adopt AdaBoost algorithm to process described pre-processed results, obtain the recognition result of described facial image.
6. a face identification device, is characterized in that, comprising:
Acquisition module, for obtaining facial image;
Profile module, for carrying out pre-service to described facial image, obtains pre-processed results; Described pre-service at least comprises: processed described facial image by skeleton pattern, determines the first elementary contour of described facial image; Described pre-processed results at least comprises described first elementary contour;
First colour of skin module, for processing described pre-processed results, obtains the recognition result of described facial image.
7., by face identification device according to claim 6, it is characterized in that, also comprise: the second colour of skin module;
Described second colour of skin module, for before determining the first elementary contour of described facial image, processes described facial image, determines the second elementary contour of described facial image; Described pre-processed results also comprises described second elementary contour.
8., by face identification device according to claim 7, it is characterized in that, also comprise: gradation processing module;
Described gradation processing module, before the first elementary contour of determining described facial image or the second elementary contour, carries out gray proces to described facial image; Described gray proces comprises one of following: to described facial image carry out gray-scale map conversion, gray-scale map histogram equalization, vertical gray-level projection, level and smooth after vertical gray-level projection process.
9. by the face identification device described in any one of claim 6-8, it is characterized in that, described profile module, specifically for adopting sobel operator to process described facial image, determining the first elementary contour of described facial image.
10. by face identification device according to claim 9, it is characterized in that, described first colour of skin module, specifically for adopting AdaBoost algorithm to process described pre-processed results, obtaining the recognition result of described facial image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410707594.XA CN104376310A (en) | 2014-11-27 | 2014-11-27 | Face recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410707594.XA CN104376310A (en) | 2014-11-27 | 2014-11-27 | Face recognition method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104376310A true CN104376310A (en) | 2015-02-25 |
Family
ID=52555208
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410707594.XA Pending CN104376310A (en) | 2014-11-27 | 2014-11-27 | Face recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104376310A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665341A (en) * | 2017-09-30 | 2018-02-06 | 珠海市魅族科技有限公司 | One kind identification control method, electronic equipment and computer product |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
| CN103049733A (en) * | 2011-10-11 | 2013-04-17 | 株式会社理光 | Human face detection method and human face detection equipment |
| US20130163829A1 (en) * | 2011-12-21 | 2013-06-27 | Electronics And Telecommunications Research Institute | System for recognizing disguised face using gabor feature and svm classifier and method thereof |
-
2014
- 2014-11-27 CN CN201410707594.XA patent/CN104376310A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
| CN103049733A (en) * | 2011-10-11 | 2013-04-17 | 株式会社理光 | Human face detection method and human face detection equipment |
| US20130163829A1 (en) * | 2011-12-21 | 2013-06-27 | Electronics And Telecommunications Research Institute | System for recognizing disguised face using gabor feature and svm classifier and method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 丁路: "基于局部纹理特征的人脸识别技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665341A (en) * | 2017-09-30 | 2018-02-06 | 珠海市魅族科技有限公司 | One kind identification control method, electronic equipment and computer product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108701216B (en) | Face recognition method and device and intelligent terminal | |
| KR102290392B1 (en) | Method and apparatus for registering face, method and apparatus for recognizing face | |
| EP4099221A1 (en) | Face recognition method and apparatus | |
| CN108182397B (en) | Multi-pose multi-scale human face verification method | |
| US9443137B2 (en) | Apparatus and method for detecting body parts | |
| CN103902977A (en) | Face identification method and device based on Gabor binary mode | |
| US10860755B2 (en) | Age modelling method | |
| CN109948399A (en) | A kind of the face method of payment and device of smart phone | |
| KR20140135078A (en) | Feature descriptor for robust facial expression recognition | |
| CN110991258B (en) | Face fusion feature extraction method and system | |
| CN113302619B (en) | System and method for evaluating target area and characteristic points | |
| CN105335685A (en) | Image identification method and apparatus | |
| CN107895154B (en) | Method and system for forming facial expression intensity calculation model | |
| CN105404883B (en) | A kind of heterogeneous three-dimensional face identification method | |
| CN112101293A (en) | Facial expression recognition method, device, equipment and storage medium | |
| CN104376310A (en) | Face recognition method and device | |
| Richtsfeld et al. | Implementation of Gestalt principles for object segmentation | |
| Vezzetti et al. | Application of geometry to rgb images for facial landmark localisation-a preliminary approach | |
| Riaz et al. | A model based approach for expressions invariant face recognition | |
| CN106682627B (en) | Palm print data identification method and device | |
| KR101756959B1 (en) | Image analyze method and apparatus thereby | |
| CN109376746A (en) | A kind of image identification method and system | |
| CN111428670B (en) | Face detection method, face detection device, storage medium and equipment | |
| CN104050457A (en) | Human face gender identification method based on small sample training library | |
| JP6467817B2 (en) | Image processing apparatus, image processing method, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150225 |
|
| WD01 | Invention patent application deemed withdrawn after publication |