CN104091159A - Human face identification method - Google Patents
Human face identification method Download PDFInfo
- Publication number
- CN104091159A CN104091159A CN201410333136.4A CN201410333136A CN104091159A CN 104091159 A CN104091159 A CN 104091159A CN 201410333136 A CN201410333136 A CN 201410333136A CN 104091159 A CN104091159 A CN 104091159A
- Authority
- CN
- China
- Prior art keywords
- face
- identification method
- human face
- face identification
- training set
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a human face identification method. The method comprises the steps that human face data information is acquired with an image acquisition device, and human face data information of multiple persons is acquired to serve as a training set to be guided into a testing device; the training set is pre-sorted and converted into a matrix form; the symmetrical nonnegative matrix factorization algorithm is applied to the training set to obtain a low-dimension subspace; finally, data information of human faces to be identified is mapped into the low-dimension subspace, and the specific sorts of the data information of the human faces to be identified are determined by means of a human face sorting method. By the adoption of the quick and precise human face identification method, the problems of a traditional human face identification method that convergence rate is low and time consumption is high are solved.
Description
Technical field
The invention belongs to digital image processing techniques field, particularly a kind of face identification method.
Background technology
Face recognition technology is a kind of recognition method based on biological characteristic, and compared with the recognition method traditional with fingerprint recognition etc., because it carries out sample collection by camera, this process is without any need for direct contact, thus have simply, advantage efficiently; And the mankind will identify each other according to people's facial information, thereby it also has intuitive; And because the collection of face does not need to need special external unit to gather as fingerprint is the same with iris, reduced from equipment and faked or the danger of leakage of information, therefore it has also reduced system by the possibility of impersonation.So face recognition technology is all widely used at numerous areas, and face characteristic in face recognition technology extracts and pattern-recognition is one of focus based on biological characteristic research in recent years.
Face recognition technology has been widely used in the fields such as government, army, bank, welfare, ecommerce, safe defence at present.Such as, a depositor goes bank debits, and he can not be with bank card, also need not recall password, because he when withdraw deposit on cash machine, a video camera scans this user's face, has then completed rapidly and exactly user identity qualification, has handled business.In addition, after U.S.'s September 11 attacks, Antiterrorism has become the common recognition of national governments, and the safe defence of strengthening the public places such as airport, market, railway station, bus station is very important.The face recognition technology of Visa lattice company of the U.S. is just given full play to one's remarkable skill on two airports of the U.S., and it can choose a certain face in crowded crowd, judges that he is wanted criminal.
Along with the further maturation of face recognition technology and the raising of Social Agree, face recognition technology will be applied in more field.Such as enterprise, house safety and management, as recognition of face access control and attendance system, recognition of face antitheft door etc.; E-Passport and I.D., International Civil Aviation Organization (ICAO) is definite, from 2010, its 118 member countries and regions, must use machine-readable passport, face recognition technology is first-elected recognition mode, and this regulation has become international standard, and the E-Passport plan Ministry of Public Security one of China is stepping up planning and implementation; Public security, the administration of justice and criminal investigation, security department can utilize face identification system and network, tracks down and arrests runaway convict in China; Information security, such as computer log, E-Government and ecommerce, current, the mandate of transaction or examination & approval is all to realize by password, if password is stolen, just cannot ensures safety, but use face recognition technology, just can accomplish that litigant is unified at online digital identity and true identity, thereby greatly increase the reliability of e-commerce and e-government system.
And from economic benefit, biometrics identification technology is in recent years with 20% to 30% speed increment every year on average, wherein, amplification is up to 80% especially for face recognition technology, and the domestic market share reaches more than 8,000 ten thousand yuan at present.So in the coming years, the market share of China's face recognition technology will reach billions of yuans.Meanwhile, face recognition technology also has huge potential market, i.e. a government market.Such as, the social security sector of China some areas is in order to ensure social security fund safety, prevent and stop to deceive neck, emit the situation such as draw one's pension to occur, having adopted one after another " face Dynamic Recognition authentication system ", to strengthen authentication and the management to retired personnel.
Except government's large-scale application market presents the situation of saving up strength to start out, face recognition products is embedded into access control product of new generation in access control system just along with place mat a few years ago and reach its maturity.Gate control system in security protection industry has become one of modal security protection subsystem in most intelligent projects, obtain extensive application in fields such as government, enterprise, factory, petrochemical industry, automobile, shipbuilding, finance, hospital, armies, meanwhile, the market demand presents the development situation of rapid growth.Related data demonstration, the security access control system product of China and the industrial chain in support equipment market have realized the growth rate in year 20% to 25% in recent years.
Can therefore, just become very important and urgent for the exploitation of face recognition technology, provide better, more stable algorithm, the innovation of then carrying out on this basis product and technology also becomes an important task in current face recognition technology market.
Summary of the invention
The problem existing in order to solve background technology, the present invention aims to provide a kind of face identification method, solves that traditional face identification method speed of convergence is slow, the problem of length consuming time.
In order to realize above-mentioned technical purpose, technical scheme of the present invention is:
A kind of face identification method,
(1) gather face data message by image capture device, the face data message that gathers multiple people imports in testing apparatus as training set;
(2) testing apparatus is presorted training set, and is converted into matrix form;
(3) symmetrical Algorithms of Non-Negative Matrix Factorization is acted on to training set, thereby the raw data dimensionality reduction of higher-dimension, obtain low n-dimensional subspace n;
(4) face data message to be identified is mapped in the low n-dimensional subspace n that step (3) obtains;
(5) utilize face classification method to determine the specific category of face data message to be identified.
Wherein, presorting in above-mentioned steps (2) is using everyone face data message as a class.
Wherein, the face classification method in above-mentioned steps (5) is 1-nearest method or k-nearest method.
Wherein, above-mentioned image capture device is digital camera.
Wherein, above-mentioned testing apparatus is PC.
The beneficial effect that adopts technique scheme to bring is:
Concrete image abstraction is changed into matrix disposal by the present invention, facilitates like this processing of data.Symmetrical nonnegative matrix algorithm is used in the middle of dimension depression of order, this algorithm not only can carry out dimensionality reduction compression to raw data fast, and can ensure the validity of the dimension extracting, so, we in the middle of recognition of face, can identify the classification of face by this algorithm application fast accurately.
Brief description of the drawings
Fig. 1 is process flow diagram of the present invention.
Embodiment
Below with reference to accompanying drawing, technical scheme of the present invention is elaborated.
Process flow diagram of the present invention as shown in Figure 1, a kind of face identification method, comprises following steps:
(1) gather face data message by image capture device, the face data message that gathers multiple people imports in testing apparatus as training set;
(2) testing apparatus is presorted training set, and is converted into matrix form, and presorting is here using everyone face data message as a class;
(3) symmetrical Algorithms of Non-Negative Matrix Factorization is acted on to training set, thereby the raw data dimensionality reduction of higher-dimension, obtain low n-dimensional subspace n;
(4) face data message to be identified is mapped in the low n-dimensional subspace n that step (3) obtains;
(5) utilize face classification method to determine the specific category of face data message to be identified, the face classification method here adopts 1-nearest method or k-nearest method.
In the present embodiment, image capture device adopts digital camera, and testing apparatus adopts PC.
Taking ORL database as example, ORL database contains 40 people, and everyone has 10 face information, selects everyone front 5 face information as training sample, and rear 5 face information are as test sample book.Training sample is imported in PC, PC machine is classified and converts symmetric matrix to each artificial class training sample, by training sample being done to symmetrical Non-negative Matrix Factorization, obtain the non-negative subspace of a low-dimensional, and then test sample book is projected to this lower dimensional space, by k-nearest classification, test sample book is carried out to Classification and Identification.The discrimination of the method approaches 90%.
Above embodiment only, for explanation technological thought of the present invention, can not limit protection scope of the present invention with this, every technological thought proposing according to the present invention, and any change of doing on technical scheme basis, within all falling into protection domain of the present invention.
Claims (5)
1. a face identification method, is characterized in that:
(1) gather multiple people's face data message by image capture device, these face data messages import in testing apparatus as training set;
(2) testing apparatus is presorted training set, and is converted into matrix form;
(3) symmetrical Algorithms of Non-Negative Matrix Factorization is acted on to training set, thereby the raw data dimensionality reduction of higher-dimension, obtain low n-dimensional subspace n;
(4) face data message to be identified is mapped in the low n-dimensional subspace n that step (3) obtains;
(5) utilize face classification method to determine the specific category of face data message to be identified.
2. a kind of face identification method according to claim 1, is characterized in that: presorting in described step (2) is using everyone face data message as a class.
3. a kind of face identification method according to claim 1, is characterized in that: the face classification method in described step (5) is 1-nearest method or k-nearest method.
4. a kind of face identification method according to claim 1, is characterized in that: described image capture device is digital camera.
5. a kind of face identification method according to claim 1, is characterized in that: described testing apparatus is PC.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410333136.4A CN104091159A (en) | 2014-07-14 | 2014-07-14 | Human face identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410333136.4A CN104091159A (en) | 2014-07-14 | 2014-07-14 | Human face identification method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104091159A true CN104091159A (en) | 2014-10-08 |
Family
ID=51638874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410333136.4A Pending CN104091159A (en) | 2014-07-14 | 2014-07-14 | Human face identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104091159A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106599785A (en) * | 2016-11-14 | 2017-04-26 | 深圳奥比中光科技有限公司 | Method and device for building human body 3D feature identity information database |
| CN108108760A (en) * | 2017-12-19 | 2018-06-01 | 山东大学 | A kind of fast human face recognition |
| CN112907775A (en) * | 2021-01-27 | 2021-06-04 | 江西中科瓦力科技有限公司 | Attendance system based on face recognition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090220127A1 (en) * | 2008-02-28 | 2009-09-03 | Honeywell International Inc. | Covariance based face association |
| CN101673348A (en) * | 2009-10-20 | 2010-03-17 | 哈尔滨工程大学 | Human face recognition method based on supervision isometric projection |
-
2014
- 2014-07-14 CN CN201410333136.4A patent/CN104091159A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090220127A1 (en) * | 2008-02-28 | 2009-09-03 | Honeywell International Inc. | Covariance based face association |
| CN101673348A (en) * | 2009-10-20 | 2010-03-17 | 哈尔滨工程大学 | Human face recognition method based on supervision isometric projection |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106599785A (en) * | 2016-11-14 | 2017-04-26 | 深圳奥比中光科技有限公司 | Method and device for building human body 3D feature identity information database |
| CN108108760A (en) * | 2017-12-19 | 2018-06-01 | 山东大学 | A kind of fast human face recognition |
| CN112907775A (en) * | 2021-01-27 | 2021-06-04 | 江西中科瓦力科技有限公司 | Attendance system based on face recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104866829B (en) | A kind of across age face verification method based on feature learning | |
| CN104751143B (en) | A kind of testimony of a witness verifying system and method based on deep learning | |
| CN103593673B (en) | A kind of on-line trial authentication method based on dynamic threshold | |
| CN102629320B (en) | Ordinal measurement statistical description face recognition method based on feature level | |
| CN103778409A (en) | Human face identification method based on human face characteristic data mining and device | |
| CN102156887A (en) | Human face recognition method based on local feature learning | |
| CN103632147A (en) | System and method for implementing standardized semantic description of facial features | |
| CN103065122A (en) | Facial expression recognition method based on facial motion unit combination features | |
| CN104700094A (en) | Face recognition method and system for intelligent robot | |
| US20190147218A1 (en) | User specific classifiers for biometric liveness detection | |
| CN101739742A (en) | Networking type multi-channel access control and attendance system | |
| CN111178136A (en) | Big data-based smart campus identity recognition method and system | |
| Ali et al. | Automated attendance management systems: systematic literature review | |
| CN113420294A (en) | Malicious code detection method based on multi-scale convolutional neural network | |
| CN106228146A (en) | A kind of subregion dynamic identifying method and system | |
| CN112052731A (en) | Intelligent portrait recognition card punching attendance system and method | |
| CN104091159A (en) | Human face identification method | |
| Sun et al. | Opportunities and challenges for biometrics | |
| CN103942545A (en) | Method and device for identifying faces based on bidirectional compressed data space dimension reduction | |
| Ni | Face recognition based on deep learning under the background of big data | |
| Boncolmo et al. | Gender Identification Using Keras Model Through Detection of Face | |
| CN103294981A (en) | Intelligent face recognition system | |
| CN102368291A (en) | Personal authentication system based on invisible consciousness of fingerprint image | |
| Abdulhak et al. | A systematic review of features identification and extraction for behavioral biometrie authentication in touchscreen mobile devices | |
| CN101639876A (en) | Identity authentication method |
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 |
Application publication date: 20141008 |
|
| RJ01 | Rejection of invention patent application after publication |