CN101739571A - Block principal component analysis-based device for confirming face - Google Patents
Block principal component analysis-based device for confirming face Download PDFInfo
- Publication number
- CN101739571A CN101739571A CN200910218071A CN200910218071A CN101739571A CN 101739571 A CN101739571 A CN 101739571A CN 200910218071 A CN200910218071 A CN 200910218071A CN 200910218071 A CN200910218071 A CN 200910218071A CN 101739571 A CN101739571 A CN 101739571A
- Authority
- CN
- China
- Prior art keywords
- face
- var
- sub
- piece
- projection vector
- 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
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to a kind of device for confirming face based on block principal component analysis, which includes: people's information database to be confirmed: being stored with the smart card information of each holder, N face the samples pictures x1, x2 of each holder,.., the projection vector of xi.., xN and each samples pictures
; For calculating the device of holder's face picture projection vector Zvar; Device for confirming face: the projection vector of samples pictures xi is calculated
Euclidean between the projection vector Zvar of face picture xvar away from D (
, Zvar), if D (
, Zvar) and the threshold value that is less than setting then thinks to hold the true owner of artificial smart card. The present invention pre-processes image using Census transform method, reduces influence of the illumination variation to confirming face. Using the algorithm based on block principal component analysis, calculation amount is smaller, and analysis identification is fast, is suitable for customs monitoring system, access control system, attendance checking system etc..
Description
Technical field
The present invention relates to a kind of device for confirming face.
Background technology
In informationalized today, society and individual have become more and more urgent to the demand of the identity identifying technology of safe ready.It mainly is by physical mediums such as key, password, personal identification proofs that traditional personal identification is differentiated, the defective of these method is to be stolen easily and to lose secret meaning.And tighter security often brings a lot of inconvenience.In this case, utilize the biological characteristic of human body self to carry out identification and become trend of the times.In numerous biological identification technologies, recognition of face does not then need the person of being identified to cooperate, and is fit to not wish the occasion of being differentiated that the people discovers.And the process of recognition of face is similar to human biology custom, be easy to be accepted by masses, so recognition of face is one of biometrics identification technology that is most widely used at present, particularly at noncontact environment and letting alone under detected person's the situation, the superiority of face recognition technology is considerably beyond detection methods such as existing iris, fingerprints.
Existing people's face is confirmed and the software of analyzing identification is based on the human face analysis and the recognizer of active shape model, active appearance models, active contour model, principal component analytical method and neural network mostly, the characteristics of these methods are that discrimination is higher, shortcoming is that algorithm is too complicated, calculated amount is huge, analysis time is long, need be based upon on the hardware system of parallel computation computing time if will obtain faster, and cost is higher.
The expression shape change of people's face also has a significant impact confirming face, and Gabor small echo commonly used etc. are eliminated the method for expression influence often calculated amount is bigger, is unsuitable for the application of systems such as gate inhibition, work attendance etc.
Summary of the invention
The technical problem to be solved in the present invention provides the device for confirming face based on block principal component analysis that a kind of cost is low, the transfer pair confirming face influence of expressing one's feelings is discerned soon and can be reduced effectively in analysis.
In order to solve the problems of the technologies described above, the device for confirming face based on block principal component analysis of the present invention comprises:
People's information database to be confirmed: store each holder's smart card information, each holder's N opens people's face samples pictures x
1, x
2..., x
i..., x
NProjection vector with each samples pictures
Projection vector wherein
Obtain according to following method:
N to each holder of data library storage opens people's face samples pictures x
1, x
2..., x
i..., x
NCarry out the Census conversion and remove illumination effect; With the every width of cloth image x after the conversion
iBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; According to the weight of each sub-piece of setting, each sub-piece is chosen the proper vector and the eigenwert of varying number, form every width of cloth image x
iTransformation matrix
Utilize transformation matrix
Calculating respective projection vector
Be used for calculating everybody the face picture projection vector Z that holds
VarDevice:
With everybody the face image data information x that holds that gathers
VarCarry out the Census conversion and remove illumination effect; With the image x after the conversion
VarBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; According to the weight of each sub-piece of setting, each sub-piece is chosen the proper vector and the eigenwert of varying number, composition diagram is as x
VarTransformation matrix
Utilize transformation matrix at last
Try to achieve x
VarProjection vector Z
Var
The device that is used for confirming face:
Calculate samples pictures x
iProjection vector
With people's face picture x
VarProjection vector Z
VarBetween Euclidean distance
If
Then think the true owner of artificial smart card of holding less than preset threshold.
The present invention uses the Census transform method that image is carried out pre-service, extracts image local architectural feature to be detected, has reduced the influence of illumination variation to confirming face.Employing is based on the algorithm of block principal component analysis, holder's samples pictures and people's face picture are divided into a plurality of sub-pieces, eyes, nose change less when changing according to human face expression, face, mouth change characteristics greatly, the weight of each sub-piece of setting, thereby the sub-piece of different human face region correspondences is extracted the proper vector of varying number, can reduce the influence of expression transfer pair confirming face effectively.Calculated amount of the present invention is less, and it is fast to analyze identification, is applicable to the customs monitoring system, gate control system, attendance checking system etc.
Description of drawings
Below in conjunction with the drawings and specific embodiments the present invention is described in further detail.
Fig. 1 is the block diagram of the device for confirming face based on block principal component analysis of the present invention.
Fig. 2 is a computing machine program flow diagram in the specific embodiment of the invention.
Fig. 3 piece of face picture and samples pictures of behaving is divided synoptic diagram.
Embodiment
As shown in Figure 1, the device for confirming face based on block principal component analysis of the present invention comprises: people's information database to be confirmed is used for calculating everybody the face picture projection vector Z that holds
VarDevice, be used for the device of confirming face.
The present invention can be by the realization that programs on computing machine, DSP.
As shown in Figure 2, described computer program flow process comprises the steps:
Be used to set up the step of people's information database to be confirmed:
The N that stores each holder opens the projection vector of people's face samples pictures, smart card information and each samples pictures
Projection vector wherein
Obtain according to following method:
Samples pictures x to the holder
1, x
2..., x
i..., x
NCarry out Census conversion (Ramin Zabih, John Woodfill, " A non-parametric approach to visual correspondence ", IEEETransactions on Pattern Analysis and Machine Intelligence, 1996) the removal illumination effect;
Utilize the principal component analysis (PCA) algorithm computation projection vector of piecemeal
With the every width of cloth image x after the conversion
iBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; As shown in Figure 3, the weight of setting black, grey, white sub-piece is respectively 1,2,4, extracts black, grey, white sub-piece preceding k, 2k and 4k proper vector and eigenwert respectively with composition x
iTransformation matrix
Calculating respective projection vector
The hold step of everybody face picture and smart card information of collection.
Be used for calculating everybody the face picture projection vector Z that holds
VarStep:
With everybody the face image data information x that holds that gathers
VarCarry out the Census conversion and remove illumination effect;
Utilize the principal component analysis (PCA) algorithm computation projection vector Z of piecemeal
Var: with the image x after the conversion
VarBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; The weight of setting black, grey, white sub-piece is respectively 1,2,4, extracts black, grey, white sub-piece preceding k, 2k and 4k proper vector and eigenwert respectively, and composition diagram is as x
VarTransformation matrix
Try to achieve x at last
VarProjection vector Z
Var
The step that is used for confirming face:
In people's information database to be confirmed, find and holder's smart card information corresponding sample picture x
iProjection vector
If
Then think the true owner of artificial smart card of holding less than preset threshold, otherwise identity is rejected; Promptly
Wherein, the selection of N, t, k does not have strict restriction, and according to the processing speed that hardware can reach, processing power is strong more, and it is big more that N, t, k value can be selected.It is 4,6 or 10 that the processing speed that the present invention can reach according to hardware can be selected N, and t can be chosen as 3 * 5, and k can be chosen as 10 or 100.Choosing of each sub-piece weight is not limited to aforesaid way, and people's face zones of different changes size and chooses in the time of can be according to expression shape change.Under the situation of not considering to express one's feelings the comparison of transfer pair people face and confirming to influence, the weight of each sub-piece can all be chosen as 1.
Claims (1)
1. device for confirming face based on block principal component analysis is characterized in that comprising:
People's information database to be confirmed: store each holder's smart card information, each holder's N opens people's face samples pictures x
1, x
2..., x
i..., x
NProjection vector with each samples pictures
Projection vector wherein
Obtain according to following method:
N to each holder of data library storage opens people's face samples pictures x
1, x
2..., x
i..., x
NCarry out the Census conversion and remove illumination effect; With the every width of cloth image x after the conversion
iBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; According to the weight of each sub-piece of setting, each sub-piece is chosen the proper vector and the eigenwert of varying number, form every width of cloth image x
iTransformation matrix
, utilize transformation matrix
Calculating respective projection vector
Be used for calculating everybody the face picture projection vector Z that holds
VarDevice:
With everybody the face image data information x that holds that gathers
VarCarry out the Census conversion and remove illumination effect; With the image x after the conversion
VarBe divided into t sub-piece, try to achieve the proper vector and the eigenwert of each sub-piece respectively; According to the weight of each sub-piece of setting, each sub-piece is chosen the proper vector and the eigenwert of varying number, composition diagram is as x
VarTransformation matrix
, utilize transformation matrix at last
Try to achieve x
VarProjection vector Z
Var
The device that is used for confirming face:
Calculate samples pictures x
iProjection vector
With people's face picture x
VarProjection vector Z
VarBetween Euclidean distance D
If D
Then think the true owner of artificial smart card of holding less than preset threshold.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910218071A CN101739571A (en) | 2009-12-22 | 2009-12-22 | Block principal component analysis-based device for confirming face |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910218071A CN101739571A (en) | 2009-12-22 | 2009-12-22 | Block principal component analysis-based device for confirming face |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101739571A true CN101739571A (en) | 2010-06-16 |
Family
ID=42463037
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200910218071A Pending CN101739571A (en) | 2009-12-22 | 2009-12-22 | Block principal component analysis-based device for confirming face |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101739571A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102163283A (en) * | 2011-05-25 | 2011-08-24 | 电子科技大学 | Method for extracting face characteristic based on local three-value mode |
CN102722866A (en) * | 2012-05-22 | 2012-10-10 | 西安电子科技大学 | Compressive sensing method based on principal component analysis |
CN108647640A (en) * | 2018-05-10 | 2018-10-12 | 王逸人 | The method and electronic equipment of recognition of face |
CN109117745A (en) * | 2018-07-23 | 2019-01-01 | 青岛理工大学 | A kind of cloud recognition of face and localization method based on Building Information Model |
-
2009
- 2009-12-22 CN CN200910218071A patent/CN101739571A/en active Pending
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102163283A (en) * | 2011-05-25 | 2011-08-24 | 电子科技大学 | Method for extracting face characteristic based on local three-value mode |
CN102163283B (en) * | 2011-05-25 | 2012-08-29 | 电子科技大学 | Method for extracting face characteristic based on local three-value mode |
CN102722866A (en) * | 2012-05-22 | 2012-10-10 | 西安电子科技大学 | Compressive sensing method based on principal component analysis |
CN108647640A (en) * | 2018-05-10 | 2018-10-12 | 王逸人 | The method and electronic equipment of recognition of face |
CN109117745A (en) * | 2018-07-23 | 2019-01-01 | 青岛理工大学 | A kind of cloud recognition of face and localization method based on Building Information Model |
CN109117745B (en) * | 2018-07-23 | 2021-11-09 | 青岛理工大学 | Cloud face recognition and positioning method based on building information model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106934359B (en) | Multi-view gait recognition method and system based on high-order tensor subspace learning | |
CN110610129A (en) | Deep learning face recognition system and method based on self-attention mechanism | |
CN107403084B (en) | Gait data-based identity recognition method | |
CN104517104A (en) | Face recognition method and face recognition system based on monitoring scene | |
CN101630364A (en) | Method for gait information processing and identity identification based on fusion feature | |
CN103902977A (en) | Face identification method and device based on Gabor binary mode | |
CN101739742A (en) | Networking type multi-channel access control and attendance system | |
CN108564040B (en) | Fingerprint activity detection method based on deep convolution characteristics | |
CN107169479A (en) | Intelligent mobile equipment sensitive data means of defence based on fingerprint authentication | |
Shamrat et al. | A deep learning approach for face detection using max pooling | |
CN103077378A (en) | Non-contact human face identifying algorithm based on expanded eight-domain local texture features and attendance system | |
Liu et al. | Fingerprint presentation attack detector using global-local model | |
Wang et al. | Fusion of LDB and HOG for Face Recognition | |
JP7141518B2 (en) | Finger vein matching method, device, computer equipment, and storage medium | |
CN103593651B (en) | Based on gait and the coal mine down-hole personnel authentication identifying method of two dimension discriminant analysis | |
CN101739571A (en) | Block principal component analysis-based device for confirming face | |
Zeng et al. | Palmprint recognition using Gabor feature-based two-directional two-dimensional linear discriminant analysis | |
Li et al. | Face liveness detection based on parallel CNN | |
CN105404883A (en) | Heterogeneous three-dimensional face recognition method | |
TW201941018A (en) | Control method of fingerprint identification module being suitable for the fingerprint identification module and a control module including a classifier | |
CN102147862A (en) | Face feature extracting method based on survival exponential entropy | |
Kalangi et al. | Deployment of Haar Cascade algorithm to detect real-time faces | |
Ruiqin et al. | Design of Face Recognition Access Entrance Guard System with Mask Based on Embedded Development | |
Sukkar et al. | A Real-time Face Recognition Based on MobileNetV2 Model | |
Lahiani et al. | Comparative study beetween hand pose estimation systems for mobile devices. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100616 |