CN106339701A - Face image recognition method and system - Google Patents
Face image recognition method and system Download PDFInfo
- Publication number
- CN106339701A CN106339701A CN201610930186.XA CN201610930186A CN106339701A CN 106339701 A CN106339701 A CN 106339701A CN 201610930186 A CN201610930186 A CN 201610930186A CN 106339701 A CN106339701 A CN 106339701A
- Authority
- CN
- China
- Prior art keywords
- identified
- facial image
- face image
- image
- registered face
- 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
-
- 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
- G06V10/30—Noise filtering
-
- 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a face image recognition method and system wherein the method comprises the following steps: acquiring a face image to be recognized and performing Gaussian difference filtering on the face image to be recognized so as to obtain a filtered face image to be recognized; carrying out LBP operations to the face image to be recognized, and obtaining a histogram from the result of the LBP operations; using the histogram matching to find out a registered face image corresponding to the face image to be recognized from a registered face image training set; calculating for an overall reconstruction coefficient dispersion degree SCI of the filtered face image to be recognized and the corresponding registered face image; and based on the SCI, determining whether the face image to be recognized is the registered face image or not. The method and system of the invention utilize the histogram matching to recognize a face image. The calculation speed is fast and the sensitivity to the pose, lighting, expression and environmental change can be reduced.
Description
Technical field
The present invention relates to field of computer technology is and in particular to a kind of facial image recognition method and system.
Background technology
The development of multimedia technology, so that the form of expressing information more horn of plenty on computers, largely changes
Computer application field.Wherein, image is a kind of more extensive media of application, in particular with image processing techniques
Development is so that image has become as computer and a kind of important information of internet arena carries form.Image recognition is
Refer to the image object carrying in identification image.
Recognition of face is a kind of biometrics identification technology based on technology such as computer, Image Processing and Pattern Recognition.
Recently, it is used widely in business and law enforcement agency with recognition of face, for example the identification of criminal identification, credit card, safety are
System, on-site supervision etc., face recognition technology is more and more more paid close attention to.
In face recognition process, the change of illumination condition is the one of the main reasons leading to face identification rate to decline.Example
As, the face registration that people is carried out indoors, can normally identify under indoor conditions, but just excessively poor in outdoor recognition effect,
Even can be because the difference of indoor and outdoor light conditions leads to confidence values very little thus cannot be carried out identifying.In existing removal people
In the method for face illumination, the lifting for discrimination in the case of sidelight photograph and shade is fruitful, but just also results in some
Ordinary person's face image feature produces the change outside expectation, so usually can reduce the discrimination under normal lighting conditions.Typically
Under meaning, image object is the geometric figure distinguishable by multiple sharpness of border, constitutes by specific distribution.To image object
Feature interpretation is based on to geometric feature interpretation.By computer image recognition technology, in conjunction with Histogram Matching, can drop
The low sensitivity to attitude, illumination, expression and environmental change, can be fast and accurately to images to be recognized and target image
Similarity is analyzed judging.
Content of the invention
The technical problem to be solved is for the deficiencies in the prior art, provides a kind of image-recognizing method and is
System.
The technical scheme is that a kind of facial image recognition method, walk including following
Rapid:
Step s1, gathers facial image to be identified, and carries out difference of Gaussian filtering process to described facial image to be identified,
Obtain filtered facial image to be identified;
Step s2, carries out lbp computing, and obtains histogram from the result of lbp computing to facial image to be identified;
Step s3, using Histogram Matching, by described filtered facial image to be identified with prestore through Gauss
Registered face training set of images after differential filtering is compared, find out from described registered face training set of images described in wait to know
The corresponding described registered face image of others' face image;
Step s4, is calculated described filtered facial image to be identified and described corresponding described registered face image
Overall reconstruction coefficients degree of scatter sci;
According to described sci, step s5, judges whether described facial image to be identified is registered face image.
The invention has the beneficial effects as follows: Applied Digital image analysis technology of the present invention, using Histogram Matching, calculating speed
Hurry up, it is possible to decrease the sensitivity to attitude, illumination, expression and environmental change.
On the basis of technique scheme, the present invention can also do following improvement.
Further, before described step s1, also include step s0, facial image to be identified is normalized.
Beneficial effect using above-mentioned further scheme is: by initializing to facial image to be identified, it is right to be easy to
Image carries out later stage process.
Further, described step s1 also includes step s11, and described facial image to be identified is divided into multiple sub-blocks.
Beneficial effect using above-mentioned further scheme is: by facial image to be identified is divided into multiple sub-blocks, just
In accurate algorithm process is carried out to facial image to be identified.
Further, in described step s11, non-overlapping between the plurality of sub-block.
Beneficial effect using above-mentioned further scheme is: is prevented to the algorithm conflict producing in sub-block processing procedure
With the useless operand that may increase.
Further, being implemented as of described step s5: when described sci value is more than predetermined value, wait to know described in determination
Others' face image is registered face image, otherwise, is non-registered face image.
Beneficial effect using above-mentioned further scheme is: by arranging different predetermined values, reaches according to varying environment
The purpose of adjustment accuracy of identification.
Another kind of technical scheme that the present invention solves above-mentioned technical problem is as follows: a kind of Face Image Recognition System, comprising:
Acquisition module, it is used for gathering facial image to be identified, and carries out difference of Gaussian to described facial image to be identified
Filtering process, obtains filtered facial image to be identified;
Processing module, it is used for carrying out lbp computing to facial image to be identified, and obtains straight from the result of lbp computing
Fang Tu;
Comparing module, it is used for utilizing Histogram Matching, by described filtered facial image to be identified with prestore
Compare through difference of Gaussian filtered registered face training set of images, find out from described registered face training set of images
The corresponding described registered face image of described facial image to be identified;
Computing module, it is used for being calculated described filtered facial image to be identified and described corresponding described registration
Overall reconstruction coefficients degree of scatter sci of facial image;
Judge module, it is used for judging whether described facial image to be identified is registered face image according to described sci.
The invention has the beneficial effects as follows: Applied Digital image analysis technology of the present invention, using Histogram Matching, calculating speed
Hurry up, it is possible to decrease the sensitivity to attitude, illumination, expression and environmental change.
On the basis of technique scheme, the present invention can also do following improvement.
Further, before described acquisition module, also include normalizing module, it is used for facial image to be identified is carried out
Normalized.
Beneficial effect using above-mentioned further scheme is: by initializing to facial image to be identified, it is right to be easy to
Image carries out later stage process.
Further, described acquisition module also includes splitting module, and it is many for being divided into described facial image to be identified
Individual sub-block.
Beneficial effect using above-mentioned further scheme is: by facial image to be identified is divided into multiple sub-blocks, just
In accurate algorithm process is carried out to facial image to be identified.
Further, in described segmentation module, non-overlapping between the plurality of sub-block.
Beneficial effect using above-mentioned further scheme is: is prevented to the algorithm conflict producing in sub-block processing procedure
With the useless operand that may increase.
Further, being implemented as of described judge module: when described sci value is more than predetermined value, treat described in determination
Identification facial image is registered face image, otherwise, is non-registered face image.
Beneficial effect using above-mentioned further scheme is: by arranging different predetermined values, reaches according to varying environment
The purpose of adjustment accuracy of identification.
Brief description
Fig. 1 is a kind of facial image recognition method flow chart of the present invention;
Fig. 2 is a kind of Face Image Recognition System schematic diagram of the present invention.
Specific embodiment
Below in conjunction with accompanying drawing, the principle of the present invention and feature are described, example is served only for explaining the present invention, and
Non- for limiting the scope of the present invention.
Fig. 1 is a kind of facial image recognition method flow chart of the present invention;
As shown in figure 1, a kind of facial image recognition method, comprise the steps:
Step s1, gathers facial image to be identified, and carries out difference of Gaussian filtering process to described facial image to be identified,
Obtain filtered facial image to be identified;
Step s2, carries out lbp computing, and obtains histogram from the result of lbp computing to facial image to be identified;
Step s3, using Histogram Matching, by described filtered facial image to be identified with prestore through Gauss
Registered face training set of images after differential filtering is compared, find out from described registered face training set of images described in wait to know
The corresponding described registered face image of others' face image;
Step s4, is calculated described filtered facial image to be identified and described corresponding described registered face image
Overall reconstruction coefficients degree of scatter sci;
According to described sci, step s5, judges whether described facial image to be identified is registered face image.
Applied Digital image analysis technology of the present invention, using Histogram Matching, calculating speed is fast, it is possible to decrease to attitude, light
According to, expression and environmental change sensitivity.
Specifically, in a particular embodiment of the present invention, before described step s1, also include step s0, to people to be identified
Face image is normalized, and by initializing to facial image to be identified, is easy to carry out later stage process to image.
Specifically, in a particular embodiment of the present invention, described step s1 also includes step s11, by described people to be identified
Face image is divided into multiple sub-blocks, by facial image to be identified is divided into multiple sub-blocks, is easy to facial image to be identified is entered
The accurate algorithm process of row.In described step s11, non-overlapping between the plurality of sub-block, it is prevented to sub-block processing procedure
The algorithm conflict of middle generation and the useless operand that may increase.
In step s2, the multiple sub-blocks in facial image to be identified in step s11 are carried out lbp computing respectively, each
Sub-block can obtain a corresponding histogram, and corresponding for all sub-blocks histogram is concatenated into a higher-dimension Nogata
Figure.
Specifically, in a particular embodiment of the present invention, being implemented as of described step s5: when described sci value is more than
During predetermined value, determine that described facial image to be identified is registered face image, otherwise, be non-registered face image, by setting
Different predetermined values, reaches the purpose adjusting accuracy of identification according to varying environment.
Fig. 2 is a kind of Face Image Recognition System schematic diagram of the present invention.
As shown in Fig. 2 a kind of Face Image Recognition System, comprising:
Acquisition module, it is used for gathering facial image to be identified, and carries out difference of Gaussian to described facial image to be identified
Filtering process, obtains filtered facial image to be identified;
Processing module, it is used for carrying out lbp computing to facial image to be identified, and obtains straight from the result of lbp computing
Fang Tu;
Comparing module, it is used for utilizing Histogram Matching, by described filtered facial image to be identified with prestore
Compare through difference of Gaussian filtered registered face training set of images, find out from described registered face training set of images
The corresponding described registered face image of described facial image to be identified;
Computing module, it is used for being calculated described filtered facial image to be identified and described corresponding described registration
Overall reconstruction coefficients degree of scatter sci of facial image;
Judge module, it is used for judging whether described facial image to be identified is registered face image according to described sci.
Applied Digital image analysis technology of the present invention, using Histogram Matching, calculating speed is fast, it is possible to decrease to attitude, light
According to, expression and environmental change sensitivity.
Specifically, in a particular embodiment of the present invention, before described acquisition module, also include normalizing module, its use
In being normalized to facial image to be identified, by initializing to facial image to be identified, it is easy to image is entered
The row later stage is processed.
Specifically, in a particular embodiment of the present invention, described acquisition module also includes splitting module, and it is used for will be described
Facial image to be identified is divided into multiple sub-blocks, by facial image to be identified is divided into multiple sub-blocks, is easy to people to be identified
Face image carries out accurate algorithm process.In described segmentation module, non-overlapping between the plurality of sub-block, it is prevented to sub-block
The algorithm conflict producing in processing procedure and the useless operand that may increase.
In comparing module, the multiple sub-blocks in facial image to be identified in segmentation module are carried out lbp computing respectively, often
Individual sub-block can obtain a corresponding histogram, and corresponding for all sub-blocks histogram is concatenated into a higher-dimension Nogata
Figure.
Specifically, in a particular embodiment of the present invention, being implemented as of described judge module: when described sci value is big
When predetermined value, determine that described facial image to be identified is registered face image, otherwise, be non-registered face image, by setting
Put different predetermined values, reach the purpose adjusting accuracy of identification according to varying environment.
In the description of this specification, reference term " embodiment one ", " embodiment two ", " example ", " specific example " or
The description of " some examples " etc. means that the concrete grammar, device or the feature that describe with reference to this embodiment or example are contained in this
In at least one bright embodiment or example.In this manual, the schematic representation of above-mentioned term is necessarily directed to
Identical embodiment or example.And, the specific features of description, method, device or feature can be real at any one or more
Apply in example or example and combine in an appropriate manner.Additionally, in the case of not conflicting, those skilled in the art can be by
The feature of the different embodiments described in this specification or example and different embodiment or example is combined and combines.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all spirit in the present invention and
Within principle, any modification, equivalent substitution and improvement made etc., should be included within the scope of the present invention.
Claims (10)
1. a kind of facial image recognition method is it is characterised in that comprise the steps:
Step s1, gathers facial image to be identified, and carries out difference of Gaussian filtering process to described facial image to be identified, obtain
Filtered facial image to be identified;
Step s2, carries out lbp computing, and obtains histogram from the result of lbp computing to facial image to be identified;
Step s3, using Histogram Matching, by described filtered facial image to be identified with prestore through difference of Gaussian
Filtered registered face training set of images is compared, and finds out described people to be identified from described registered face training set of images
The corresponding described registered face image of face image;
Step s4, is calculated the total of described filtered facial image to be identified and described corresponding described registered face image
Body reconstruction coefficients degree of scatter sci;
According to described sci, step s5, judges whether described facial image to be identified is registered face image.
2. a kind of facial image recognition method according to claim 1 is it is characterised in that before described step s1, also wrap
Include step s0, facial image to be identified is normalized.
3. a kind of facial image recognition method according to claim 1 is it is characterised in that described step s1 also includes step
S11, described facial image to be identified is divided into multiple sub-blocks.
4. a kind of facial image recognition method according to claim 3 is it is characterised in that in described step s11, described many
Non-overlapping between individual sub-block.
5. a kind of facial image recognition method according to claim 1 is it is characterised in that the implementing of described step s5
For: when described sci value is more than predetermined value, determines that described facial image to be identified is registered face image, otherwise, be non-registered
Facial image.
6. a kind of Face Image Recognition System is it is characterised in that include:
Acquisition module, it is used for gathering facial image to be identified, and carries out difference of Gaussian filtering to described facial image to be identified
Process, obtain filtered facial image to be identified;
Processing module, it is used for carrying out lbp computing to facial image to be identified, and obtains histogram from the result of lbp computing;
Comparing module, it is used for utilizing Histogram Matching, by described filtered facial image to be identified and the warp prestoring
Difference of Gaussian filtered registered face training set of images is compared, and finds out described from described registered face training set of images
The corresponding described registered face image of facial image to be identified;
Computing module, it is used for being calculated described filtered facial image to be identified and described corresponding described registered face
Overall reconstruction coefficients degree of scatter sci of image;
Judge module, it is used for judging whether described facial image to be identified is registered face image according to described sci.
7. a kind of Face Image Recognition System according to claim 6 is it is characterised in that before described acquisition module, go back
Including normalization module, it is used for facial image to be identified is normalized.
8. a kind of Face Image Recognition System according to claim 6 is it is characterised in that described acquisition module also includes point
Cut module, it is used for for described facial image to be identified being divided into multiple sub-blocks.
9. a kind of Face Image Recognition System according to claim 8 is it is characterised in that in described segmentation module, described
Non-overlapping between multiple sub-blocks.
10. a kind of Face Image Recognition System according to claim 6 it is characterised in that described judge module concrete
It is embodied as: when described sci value is more than predetermined value, determines that described facial image to be identified is registered face image, otherwise, be non-
Registered face image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610930186.XA CN106339701A (en) | 2016-10-31 | 2016-10-31 | Face image recognition method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610930186.XA CN106339701A (en) | 2016-10-31 | 2016-10-31 | Face image recognition method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106339701A true CN106339701A (en) | 2017-01-18 |
Family
ID=57840602
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610930186.XA Pending CN106339701A (en) | 2016-10-31 | 2016-10-31 | Face image recognition method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106339701A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107133590A (en) * | 2017-05-04 | 2017-09-05 | 上海博历机械科技有限公司 | A kind of identification system based on facial image |
CN108596057A (en) * | 2018-04-11 | 2018-09-28 | 重庆第二师范学院 | A kind of Information Security Management System based on recognition of face |
CN109271997A (en) * | 2018-08-28 | 2019-01-25 | 河南科技大学 | A kind of image texture classification method based on jump subdivision local mode |
CN112633250A (en) * | 2021-01-05 | 2021-04-09 | 北京经纬信息技术有限公司 | Face recognition detection experimental method and device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1790374A (en) * | 2004-12-14 | 2006-06-21 | 中国科学院计算技术研究所 | Face recognition method based on template matching |
CN102262723A (en) * | 2010-05-24 | 2011-11-30 | 汉王科技股份有限公司 | Face recognition method and device |
CN103745206A (en) * | 2014-01-27 | 2014-04-23 | 中国科学院深圳先进技术研究院 | Human face identification method and system |
CN104584034A (en) * | 2012-08-15 | 2015-04-29 | 高通股份有限公司 | Method and apparatus for facial recognition |
CN104751108A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Face image recognition device and face image recognition method |
US20150347820A1 (en) * | 2014-05-27 | 2015-12-03 | Beijing Kuangshi Technology Co., Ltd. | Learning Deep Face Representation |
-
2016
- 2016-10-31 CN CN201610930186.XA patent/CN106339701A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1790374A (en) * | 2004-12-14 | 2006-06-21 | 中国科学院计算技术研究所 | Face recognition method based on template matching |
CN102262723A (en) * | 2010-05-24 | 2011-11-30 | 汉王科技股份有限公司 | Face recognition method and device |
CN104584034A (en) * | 2012-08-15 | 2015-04-29 | 高通股份有限公司 | Method and apparatus for facial recognition |
CN104751108A (en) * | 2013-12-31 | 2015-07-01 | 汉王科技股份有限公司 | Face image recognition device and face image recognition method |
CN103745206A (en) * | 2014-01-27 | 2014-04-23 | 中国科学院深圳先进技术研究院 | Human face identification method and system |
US20150347820A1 (en) * | 2014-05-27 | 2015-12-03 | Beijing Kuangshi Technology Co., Ltd. | Learning Deep Face Representation |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107133590A (en) * | 2017-05-04 | 2017-09-05 | 上海博历机械科技有限公司 | A kind of identification system based on facial image |
CN107133590B (en) * | 2017-05-04 | 2018-09-21 | 上海天使印记信息科技有限公司 | A kind of identification system based on facial image |
CN108596057A (en) * | 2018-04-11 | 2018-09-28 | 重庆第二师范学院 | A kind of Information Security Management System based on recognition of face |
CN108596057B (en) * | 2018-04-11 | 2022-04-05 | 重庆第二师范学院 | Information security management system based on face recognition |
CN109271997A (en) * | 2018-08-28 | 2019-01-25 | 河南科技大学 | A kind of image texture classification method based on jump subdivision local mode |
CN109271997B (en) * | 2018-08-28 | 2022-01-28 | 河南科技大学 | Image texture classification method based on skip subdivision local mode |
CN112633250A (en) * | 2021-01-05 | 2021-04-09 | 北京经纬信息技术有限公司 | Face recognition detection experimental method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108009472B (en) | Finger back joint print recognition method based on convolutional neural network and Bayes classifier | |
CN104036278B (en) | The extracting method of face algorithm standard rules face image | |
WO2015096565A1 (en) | Method and device for identifying target object in image | |
CN106339701A (en) | Face image recognition method and system | |
Sugiarto et al. | Wood identification based on histogram of oriented gradient (HOG) feature and support vector machine (SVM) classifier | |
CN103984951B (en) | A kind of Magnetic testing defect inspection method and system | |
US9607211B2 (en) | Face recognition apparatus, system and method for managing users based on user grouping | |
CN105389593A (en) | Image object recognition method based on SURF | |
CN108171223A (en) | A kind of face identification method and system based on multi-model multichannel | |
CN105760858A (en) | Pedestrian detection method and apparatus based on Haar-like intermediate layer filtering features | |
CN101799875B (en) | Target detection method | |
CN100370475C (en) | Method for filtering sensing images based on heteropic quantized color feature vectors | |
CN114863189B (en) | Intelligent image identification method based on big data | |
Francis et al. | TEDLESS–Text detection using least-square SVM from natural scene | |
WO2017054276A1 (en) | Biometric identity verification method and device | |
Ganapathy et al. | A Malaysian vehicle license plate localization and recognition system | |
CN105550646B (en) | Generalized illumination invariant face feature description method based on logarithmic gradient histogram | |
CN105469099B (en) | Pavement crack detection and identification method based on sparse representation classification | |
CN108122000A (en) | A kind of automobile logo identification method of feature based study | |
CN116468935A (en) | Multi-core convolutional network-based stepwise classification and identification method for traffic signs | |
Zhang et al. | Eye Location Based on Adaboost and Random Forests. | |
Tariq et al. | An automated system for fingerprint classification using singular points for biometric security | |
Adam et al. | Eyelid Localization for Iris Identification. | |
Abd Almisreb et al. | Enhancement of iris boundary detection based on Otsu method | |
CN117173494B (en) | Noise-containing label image recognition method and system based on class balance sample selection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170118 |