CN103514442B - Video sequence face identification method based on AAM model - Google Patents

Abstract

The invention discloses a video sequence face identification method based on an AAM model. The method comprises the training step (1) and the identification step (2), wherein the training step (1) comprises PCA protection and (1-2) LDA protection, projection is carried out on feature vectors which are subjected to dimensionality reduction of the PCA protection through a WLDA, and therefore the best classification feature of each training picture is obtained; the identification step (2) comprises (2-1) Adaboost detection, (2-2) AAM tracking and gesture correction, (2-3) PCA projection, (2-4) LDA protection aiming to obtain the best classification feature of a face picture to be identified, (2-5) nearest-neighbor classifier decision which judges the face picture of types where gamma1 is located to the be the identification result, and the gamma1 is the smallest euclidean distance between the best classification feature of the face picture to be identified and the best classification feature of all training pictures. The video sequence face identification method has the advantages of being capable of accurately identifying a face under the condition that the face gesture is variable, and is strong in robustness.

Description

A kind of video sequence face identification method based on AAM model

Technical field

The present invention relates to a kind of face identification method, particularly to a kind of video sequence recognition of face based on AAM model Method.

Background technology

The information age maked rapid progress in this information expansion, computer technology, the mankind start to wish that computer becomes a kind of Can with natural language between the machine that exchanged, and thirst for developing man machine interface and artificial intelligence's skill of novel concept Art, thus enable people to eliminate the reliance on the interactive devices such as keyboard, mouse and the display device of traditional computer.However, it is real Now so naturally man-machine interaction requires that computer can quickly and accurately obtain identity, state, intention and the phase of user The characteristic information closing.Because the bulk information content that face is contained is an important information transmission window, computer passes through Obtain identity and the relevant information of object using the uniqueness of face face, pass through the expression shape change that face enriches simultaneously The state of conveyed object and intention, make one to build up an intelligentized bridge and computer between, and these must have been studied The image processing techniquess related to face of effect.

At present in existing face identification method, face feature extraction method is mainly based upon geometric properties, based on template Coupling, based on subspace with based on methods such as neutral nets.In the extracting method based on subspace, principal component analysiss PCA and The methods such as Fisher linear discriminant are the more commonly used methods, and they obtain higher discrimination in still image.But In video sequence, current face identification method must could obtain preferable recognition effect in the case of user's cooperation, if In the case of user is ill-matched in identification process, recognition effect is likely to occur and is greatly reduced.In addition in the video sequence, by Changeable etc. in face posture leads to the discrimination of these methods to have different degrees of decline.

Movable appearance model (Active Appearance Model, AAM) is be widely used in area of pattern recognition one Plant Feature Points Extraction.Human face characteristic positioning method based on AAM, during setting up faceform, not only considers that local is special Reference ceases, and considers global shape and texture information, by uniting to face shape feature and textural characteristics Meter analysis, sets up face mixed model, i.e. final corresponding AAM model.

Content of the invention

It is an object of the invention to overcoming shortcoming and the deficiency of prior art, provide a kind of video sequence based on AAM model Row face identification method.The method also accurately can identify face in the case that face posture is changeable, has very strong Robustness.

The purpose of the present invention is achieved through the following technical solutions：A kind of video sequence recognition of face side based on AAM model Method, including training stage and cognitive phase；

（1）The described training stage includes：

（1-1）PCA projects：

First training picture is normalized, average face is calculated according to the training picture after normalization, by all normalizings Training picture after change and average face do difference operation, obtain the first difference；

Then according to first difference build covariance matrix, by the feature of K eigenvalue of maximum before covariance matrix to Amount composition PCA projection matrix WPCA, as feature quantity space；

Finally the first difference is projected to lower dimensional space by PCA projection matrix WPCA, obtain the characteristic vector after dimensionality reduction；

（1-2）LDA projects：

Calculate first the mean vector m of characteristic vector after the PCA dimensionality reduction that obtains of projection for all training picture samples with And i-th class train the mean vector mi of characteristic vector after the PCA dimensionality reduction that obtains of projection for the picture sample；

Then according to mean vector m, mi and calculating training sample within class scatter matrix S_WWith inter _ class relationship matrix S_B, Calculating matrix S_W ^-1S_BCharacteristic vector, by choose S_W ^-1S_BFirst L maximum characteristic vector constitute LDA projection matrix W_LDA；

Finally by LDA projection matrix W_LDAProject to projecting the characteristic vector after dimensionality reduction by PCA, obtain every Optimal classification feature Γ of training picture_ij；

（2）Described cognitive phase includes：

（2-1）Adaboost detects：The subregion that test video frame comprises face is identified by Adaboost algorithm；

（2-2）AAM follows the tracks of and posture correction：Training obtains AAM model first；Then pass through the AAM model that training obtains Face subregion is tracked；The net shape parameter obtaining when finally using AAM model training is entered to the subregion of face Row posture corrects, and obtains the face subregion after posture correction；

（2-3）PCA projects：

First the face subregion picture after posture obtained above correction is normalized, then with the training stage The average face obtaining during PCA projection does difference operation, obtains the second difference；

Then above-mentioned second matrix of differences is projected to the PCA projection matrix W that the training stage obtains_PCA, after obtaining dimensionality reduction Characteristic vector η；

（2-4）LDA projects：By step（2-3）In eigenvector projection after the dimensionality reduction that obtains obtain to the training stage LDA projection matrix W_LDA, obtain the optimal classification feature of facial image to be identified；

（2-5）Nearest neighbor classifier decision-making：

Calculate European between each training picture optimal classification feature and other training picture optimal classification features first Distance, therefrom selects Euclidean distance value F of maximum；Set threshold value b, the size of this threshold value b is maximum Euclidean distance The half of value F；

Then calculation procedure（2-4）The optimal classification feature of the facial image to be identified obtaining is each with what the training stage obtained Minimum euclid distance γ of the optimal classification feature of individual training picture₁；

Finally by minimum euclid distance γ₁Being compared with threshold value b, if being more than threshold value b, judging this people to be identified Face image is non-training storehouse picture；If being less than threshold value b, by minimum euclid distance γ₁Place characteristic of division place class Face picture is judged to recognition result.

Preferably, described step（1-1）In calculated average face f be：

$f=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{x}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Wherein x_ijFor the training picture after normalization, C is the classification sum of the training picture after normalization, and N is every apoplexy due to endogenous wind The training picture sample sum comprising, M is training picture sample sum, wherein M=N*C；

Training picture after normalization and average face do difference operation, the first difference d obtaining_ijFor：

d_ij=x_ij- f, i=1,2 ..., C, j=1,2 ... N；

According to the first matrix of differences d_ijBuild covariance matrix U be：

$U=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{d}_{\mathrm{ij}}{d_{\mathrm{ij}}}^T,i=\mathrm{1,2},...,C,j=1,2,...N;$

Wherein the first difference is passed through PCA projection matrix W_PCAProject to lower dimensional space, obtain characteristic vector η after dimensionality reduction_ij For：

η_ij=W_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N.

Further, described step（1-2）In all training picture samples through PCA projection after characteristic vector averages Vectorial m is：

$m=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\η}}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

I-th class trains the mean vector mi of characteristic vector after PCA projection for the picture sample to be：

$m_i=\frac{1}{N}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\η}}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Described step trains the within class scatter matrix S of picture sample_WWith inter _ class relationship matrix S_BIt is respectively：

$S_W=\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\η}}_{\mathrm{ij}}-m_i){({\mathrm{\η}}_{\mathrm{ij}}-m_i)}^T,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

$S_B=\underset{i=1}{\overset{C}{\mathrm{\Σ}}}{n}_i(m_i-m){(m_i-m)}^T,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Wherein ni is the number of the i-th class training sample；Optimal classification feature Γ of every training picture_ijFor：

Γ_ij=W_LDA ^TW_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N.

Preferably, described step（2-2）Middle AAM follows the tracks of as follows with the training step of the AAM model of posture correction：

（2-2-1）Choosing training object is to include positive face, left and right sides face, face of facing upward, S of face of bowing reliability sample；

（2-2-2）Described point is carried out to reliable sample, 68 feature visibility points of face are demarcated；

（2-2-3）Using Procrustes, the face after described point is alignd, obtain removing translation, yardstick and rotation Alignment face；

（2-2-4）Using principal component analytical method to step（2-2-3）The alignment face obtaining carries out shape modeling, obtains Form parameter p（I.e. coefficient of torsion）And shape；

（2-2-5）Remove average shape face from shape, then delaunay triangle division is carried out to it, then use The affine method of burst makes texture project in average shape, is finally processed with PCA, obtains parametric texture and texture mould Type；

（2-2-6）According to shape obtained above and texture model, right respectively using reversely combined aam matching algorithm Existing shape and texture model are trained, and obtain hessian matrix.

Further, described step（2-2-1）Quantity S of middle reliable sample is 100～1000.

Further, described step（2-2）The process that middle face subregion is tracked is as follows：

（2-2-7）Form parameter p according to hessian matrix and shape obtains shape by following burst mapping function Shape parameter increase △ p：

$\mathrm{\Δp}=H^{-1}\underset{x}{\mathrm{\Σ}}{[\▿I\frac{\∂W}{\∂p}]}^T[T\left(x\right)-I\left(W(x;p)\right)];$

Wherein H is hessian matrix, and W is the affine equation of burst, and T is alignment face, and I is actual picture, and x is actual figure Pixel in picture, p is the form parameter of shape in corresponding training process；

（2-2-8）Form parameter p, wherein p=p+ Δ p are updated according to form parameter increment Delta p；It is then back to step（2-2- 7）, continue to calculate Δ p by above-mentioned burst mapping function, until calculated form parameter increment Delta p is less than a or iteration Number of times reaches maximum iteration time, then stop calculating；

（2-2-9）Form parameter p obtaining after updating combines shape, and application principal component analytical method is followed the tracks of Target facial image.

Further, described threshold value a is 500 to 2000.

Further, described step（2-2）In posture trimming process as follows：When accurately tracing into target facial image Afterwards, posture correction is carried out to the subregion of face according to net shape parameter p obtaining in tracking iterative process, after being reversed Shape I (W (x;P)) it is the reconstruction face consistent with positive face.

Preferably, described step（2-3）In obtain characteristic vector η after dimensionality reduction and be：

η=W_PCA ^Tu；

Optimal classification feature Γ of facial image to be identified is：

Γ=W_LDA ^TW_PCA ^Tu；

Wherein u is step（2-3）In the second difference of obtaining；Wherein u=x-f；X is step（2-3）In the posture school that obtains The picture after face subregion normalization after just, f is step（1-1）In the average face that obtains.

Preferably, described step（2-5）In optimal classification feature Γ and training stage of facial image to be identified obtain Optimal classification feature Γ of each training picture_ijMinimum euclid distance γ₁For：

${\mathrm{\γ}}_1=\underset{i,j}{\min }\left|\right|\mathrm{\Γ}-{\mathrm{\Γ}}_{\mathrm{ij}}\left|\right|,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Wherein C is the classification sum of the training picture after normalization, and N is the training picture sample sum that every apoplexy due to endogenous wind comprises.

The present invention has such advantages as with respect to prior art and effect：

After the present inventor's face recognition method is tracked using AAM model in identification process, carries out attitude updating, obtain To a reconstruction face consistent with positive face as the face subregion after attitude updating, this process will test facial image distortion It is corrected to the facial image just to photographic head, then adopt PCA projection and LDA projection preferably to extract facial image to be identified Optimal classification feature is so that the inventive method also can accurately identify face in the case of human face posture is diverse.Tool There is strong robustness.

Brief description

Fig. 1 is the flow chart of the inventive method.

Fig. 2 is 68 feature locations of face in the inventive method.

Specific embodiment

With reference to embodiment and accompanying drawing, the present invention is described in further detail, but embodiments of the present invention do not limit In this.

Embodiment

As shown in figure 1, present embodiment discloses a kind of video sequence face identification method based on AAM model, including instruction Practice step and identification step；

（1）Training step includes：

（1-1）PCA projects：

First pretreatment is carried out to training picture, including being converted into gray scale, every training picture pulls into column vector by row, enters Row normalization, according to the training picture x after normalization_ijCalculate average face f, by the training picture x after all normalization_ijWith flat All face f do difference operation, obtain the first difference d_ij；

Wherein average face f is：

$f=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{x}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

x_ijFor j-th training picture of the i-th class after normalization, C is the classification sum of the training picture after normalization, and N is The training picture sample sum that every apoplexy due to endogenous wind comprises, M is training picture sample sum, wherein M=N*C；

First difference d_ijFor：

d_ij=x_ij- f, i=1,2 ..., C, j=1,2 ... N；

Then according to the first difference d_ijBuild covariance matrix U, accounting for before gross energy more than 95% by covariance matrix The characteristic vector composition PCA projection matrix W of K eigenvalue of maximum_PCA, as feature quantity space；The covariance matrix of wherein structure U is：

$U=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{d}_{\mathrm{ij}}{d_{\mathrm{ij}}}^T,i=\mathrm{1,2},...,C,j=1,2,...N;$

Finally the first difference is passed through PCA projection matrix W_PCAProject to lower dimensional space, obtain characteristic vector η after dimensionality reduction_ij For：

η_ij=W_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N；

（1-2）LDA projects：

Calculate first the mean vector m of characteristic vector after the PCA dimensionality reduction that obtains of projection for all training picture samples with And i-th class train the mean vector mi of characteristic vector after the PCA dimensionality reduction that obtains of projection for the picture sample；

$m=\frac{1}{M}\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\η}}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

$m_i=\frac{1}{N}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\η}}_{\mathrm{ij}},i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Then according to mean vector m, mi and calculating training sample within class scatter matrix S_WWith inter _ class relationship matrix S_B, Calculating matrix S_W ^-1S_BCharacteristic vector, by choose S_W ^-1S_BFirst L maximum characteristic vector structure accounting for gross energy more than 95% Become LDA projection matrix W_LDA；Wherein training sample within class scatter matrix S_WWith inter _ class relationship matrix S_BIt is respectively：

$S_W=\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\η}}_{\mathrm{ij}}-m_i){({\mathrm{\η}}_{\mathrm{ij}}-m_i)}^T,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

$S_B=\underset{i=1}{\overset{C}{\mathrm{\Σ}}}{n}_i(m_i-m){(m_i-m)}^T,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Wherein ni is the number of the i-th class training sample；

Finally by LDA projection matrix W_LDAProject to projecting the characteristic vector after dimensionality reduction by PCA, obtain every Optimal classification feature Γ of training picture_ij；Wherein optimal classification feature Γ_ijFor：

Γ_ij=W_LDA ^TW_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N；

（2）Described identification step includes：

（2-1）Adaboost detects：The subregion that test video frame comprises face is identified by Adaboost algorithm；

（2-2）AAM follows the tracks of and posture correction：Training obtains AAM model first；Then pass through the AAM model that training obtains Face subregion is tracked；The net shape parameter obtaining when finally using AAM model training is entered to the subregion of face Row posture corrects, and obtains the face subregion after posture correction；

Wherein AAM follows the tracks of as follows with the training step of the AAM model of posture correction：

（2-2-1）Choosing training object is to include positive face, left and right sides face, face of facing upward, S of face of bowing reliability sample, In embodiment, quantity S of reliable sample is 100～1000.

（2-2-2）Described point is carried out to reliable sample, 68 feature visibility points of face are demarcated；In Fig. 2 Shown.

（2-2-3）Using Procrustes（Pu Shi analyzes）Face after described point is alignd, obtains removing translation, chi Degree and the alignment face of rotation；

（2-2-4）Using principal component analytical method to step（2-2-3）The alignment face obtaining carries out shape modeling, obtains Form parameter p（I.e. coefficient of torsion）And shape；

（2-2-5）Remove average shape face from shape, then delaunay triangle division is carried out to it, then use The affine method of burst makes texture project in average shape, is finally processed with PCA, obtains parametric texture and texture mould Type；

（2-2-6）According to shape obtained above and texture model, right respectively using reversely combined aam matching algorithm Existing shape and texture model are trained, and obtain hessian matrix.

The process that face subregion is tracked is as follows：

（2-2-7）Form parameter p according to hessian matrix and shape obtains shape by following burst mapping function Shape parameter increase Δ p：

$\mathrm{\Δp}=H^{-1}\underset{x}{\mathrm{\Σ}}{[\▿I\frac{\∂W}{\∂p}]}^T[T\left(x\right)-I\left(W(x;p)\right)];$

Wherein, H is hessian matrix, and W is the affine equation of burst, and T is alignment face, and I is actual picture, and x is actual figure Pixel in picture, p is the form parameter of shape in corresponding training process；

（2-2-8）Form parameter p, wherein p=p+ Δ p are updated according to form parameter increment Delta p；It is then back to step（2-2- 7）, continue to calculate Δ p by above-mentioned burst mapping function, until calculated form parameter increment Delta p is less than a or iteration Number of times reaches maximum iteration time, then stop calculating；

（2-2-9）Form parameter p obtaining after updating combines shape, and application principal component analytical method is followed the tracks of Target facial image.

Posture trimming process is as follows：After accurately tracing into target facial image, according to obtain in tracking iterative process Net shape parameter p carries out posture correction to the subregion of face, the shape I (W (x after being reversed;P)) it is and positive face one The reconstruction face causing, that is, obtain the face subregion after posture correction；

（2-3）PCA projects：

First the face subregion picture after posture obtained above correction is pulled into by row and is normalized after column vector, The average face obtaining when then projecting with the PCA of training stage does difference operation, obtains the second difference；

Then above-mentioned second matrix of differences is projected to the PCA projection matrix W that the training stage obtains_PCA, after obtaining dimensionality reduction Characteristic vector η is：

η=W_PCA ^Tu；

（2-4）LDA projects：By step（2-3）In eigenvector projection after the dimensionality reduction that obtains obtain to the training stage LDA projection matrix W_LDA, optimal classification feature Γ obtaining facial image to be identified is：

Γ=W_LDA ^TW_PCA ^Tu；

Wherein u is step（2-3）In the second difference of obtaining, u=x-f；X is step（2-3）In obtain posture correction after Face subregion normalization after picture, f be step（1-1）In the average face that obtains.

（2-5）Nearest neighbor classifier decision-making：

Calculate each training picture optimal classification feature in training picture library first and train picture optimal classification feature with other Between Euclidean distance, therefrom select maximum Euclidean distance value F；Set threshold value b, the size of this threshold value b is maximum Euclidean distance value F half.

Then calculation procedure（2-4）The optimal classification feature of the facial image to be identified obtaining is each with what the training stage obtained Minimum euclid distance γ of the optimal classification feature of individual training picture₁：

${\mathrm{\γ}}_1=\underset{i,j}{\min }\left|\right|\mathrm{\Γ}-{\mathrm{\Γ}}_{\mathrm{ij}}\left|\right|,i=\mathrm{1,2},...,C,j=\mathrm{1,2},...N;$

Finally by minimum euclid distance γ₁Being compared with threshold value b, if being more than threshold value b, judging this people to be identified Face image is non-training storehouse picture；If being less than threshold value b, by minimum euclid distance γ₁Place characteristic of division place class Face picture is judged to recognition result.

Above-described embodiment is the present invention preferably embodiment, but embodiments of the present invention are not subject to above-described embodiment Limit, other any spirit without departing from the present invention and the change made under principle, modification, replacement, combine, simplify, All should be equivalent substitute mode, be included within protection scope of the present invention.

Claims (10)

1. a kind of video sequence face identification method based on AAM model, including training stage and cognitive phase；

(1) the described training stage includes：

(1-1) PCA projection：

First training picture is normalized, average face is calculated according to the training picture after normalization, after all normalization Training picture and average face do difference operation, obtain the first difference；

Then covariance matrix is built according to the first difference, by the characteristic vector group of K eigenvalue of maximum before covariance matrix Become PCA projection matrix W_PCA, as feature quantity space；

Finally the first difference is passed through PCA projection matrix W_PCAProject to lower dimensional space, obtain the characteristic vector after dimensionality reduction；

(1-2) LDA projection：

Calculate the mean vector m and i-th of characteristic vector after the dimensionality reduction that PCA projection obtains for all training picture samples first Class trains the mean vector mi of characteristic vector after the dimensionality reduction that PCA projection obtains for the picture sample；

Then according to mean vector m, mi and calculating training sample within class scatter matrix S_WWith inter _ class relationship matrix S_B, calculate Matrix S_W ^-1S_BCharacteristic vector, by choose S_W ^-1S_BFirst L maximum characteristic vector constitute LDA projection matrix W_LDA；

Finally by LDA projection matrix W_LDAProject to projecting the characteristic vector after dimensionality reduction by PCA, obtain every training Optimal classification feature Γ of picture_ij；

(2) described cognitive phase includes：

(2-1) Adaboost detection：The subregion that test video frame comprises face is identified by Adaboost algorithm；

(2-2) AAM follows the tracks of and posture correction：Training obtains AAM model first；Then pass through to train the AAM model obtaining to people Face region is tracked；The net shape parameter obtaining when finally using AAM model training carries out appearance to the subregion of face Gesture corrects, and obtains the face subregion after posture correction；

(2-3) PCA projection：

First the face subregion picture after posture obtained above correction is normalized, the then PCA with the training stage The average face obtaining during projection does difference operation, obtains the second difference；

Then above-mentioned second matrix of differences is projected to the PCA projection matrix W that the training stage obtains_PCA, obtain the feature after dimensionality reduction Vectorial η；

(2-4) LDA projection：The LDA that eigenvector projection after the dimensionality reduction that will obtain in step (2-3) obtained to the training stage throws Shadow matrix W_LDA, obtain the optimal classification feature of facial image to be identified；

(2-5) nearest neighbor classifier decision-making：

Calculate the Euclidean distance between each training picture optimal classification feature and other training picture optimal classification features first, Therefrom select Euclidean distance value F of maximum；Set threshold value b, the size of this threshold value b is maximum Euclidean distance value F Half；

Then each instruction that the optimal classification feature of the facial image to be identified that calculation procedure (2-4) obtains and training stage obtain Practice minimum euclid distance γ of the optimal classification feature of picture₁；

Finally by minimum euclid distance γ₁Being compared with threshold value b, if being more than threshold value b, judging this face figure to be identified Picture is non-training storehouse picture；If being less than threshold value b, by minimum euclid distance γ₁The face of place characteristic of division place class Picture is judged to recognition result.

2. the video sequence face identification method based on AAM model according to claim 1 is it is characterised in that described step Suddenly in (1-1), calculated average face f is：

$f=\frac{1}{M}\underset{i=1}{\overset{C}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}{x}_{ij},i=1,2,...,C,j=1,2,...N;$

Wherein x_ijFor the training picture after normalization, C is the classification sum of the training picture after normalization, and N comprises for every apoplexy due to endogenous wind Training picture sample sum, M be training picture sample sum, wherein M=N*C；

Training picture after normalization and average face do difference operation, the first difference d obtaining_ijFor：

d_ij=x_ij- f, i=1,2 ..., C, j=1,2 ... N；

According to the first matrix of differences d_ijBuild covariance matrix U be：

$U=\frac{1}{M}\underset{i=1}{\overset{C}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}{d}_{ij}{d_{ij}}^T,i=1,2,...,C,j=1,2,...N;$

Wherein the first difference is passed through PCA projection matrix W_PCAProject to lower dimensional space, obtain characteristic vector η after dimensionality reduction_ijFor：

η_ij=W_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N.

3. the video sequence face identification method based on AAM model according to claim 2 is it is characterised in that described step Suddenly in (1-2), the mean vector m of characteristic vector after the dimensionality reduction that PCA projection obtains for all training picture samples is：

$m=\frac{1}{M}\underset{i=1}{\overset{C}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}{\mathrm{\η}}_{ij},i=1,2,...,C,j=1,2,...N;$

I-th class trains the mean vector mi of characteristic vector after the PCA dimensionality reduction that obtains of projection for the picture sample to be：

$m_i=\frac{1}{N}\underset{j=1}{\overset{N}{\Σ}}{\mathrm{\η}}_{ij},i=1,2,...,C,j=1,2,...N;$

Described step trains the within class scatter matrix S of picture sample_WWith inter _ class relationship matrix S_BIt is respectively：

$S_W=\underset{i=1}{\overset{C}{\Σ}}\underset{j=1}{\overset{N}{\Σ}}({\mathrm{\η}}_{ij}-m_i){({\mathrm{\η}}_{ij}-m_i)}^T,i=1,2,...,C,j=1,2,...N;$

$S_B=\underset{i=1}{\overset{C}{\mathrm{\Σ}}}{n}_i(m_i-m){(m_i-m)}^T,i=1,2,...,C,j=1,2,...N;$

Wherein ni is the number of the i-th class training sample；Optimal classification feature Γ of every training picture_ijFor：

Γ_ij=W_LDA ^TW_PCA ^Td_ij, i=1,2 ..., C, j=1,2 ... N.

4. the video sequence face identification method based on AAM model according to claim 1 is it is characterised in that described step Suddenly in (2-2), AAM tracking is as follows with the training step of the AAM model of posture correction：

(2-2-1) choosing training object is to include positive face, left and right sides face, face of facing upward, S of face of bowing reliability sample；

(2-2-2) described point is carried out to reliable sample, 68 feature visibility points of face are demarcated；

(2-2-3) using Procrustes, the face after described point is alignd, obtain removing the alignment of translation, yardstick and rotation Face；

(2-2-4) shape modeling is carried out using the alignment face that principal component analytical method obtains to step (2-2-3), obtain shape Parameter p and shape；

(2-2-5) remove average shape face from shape, then delaunay triangle division is carried out to it, then use burst Affine method makes texture project in average shape, is finally processed with PCA, obtains parametric texture and texture model；

(2-2-6) according to shape obtained above and texture model, using reversely combined aam matching algorithm respectively to existing Shape and texture model are trained, and obtain hessian matrix.

5. the video sequence face identification method based on AAM model according to claim 4 is it is characterised in that described step Suddenly in (2-2-1), quantity S of reliable sample is 100～1000.

6. the video sequence face identification method based on AAM model according to claim 4 is it is characterised in that described step Suddenly the process that in (2-2), face subregion is tracked is as follows：

(2-2-7) form parameter p according to hessian matrix and shape obtains shape ginseng by following burst mapping function Number increment Delta p：

$\mathrm{\Δ}p=H^{-1}\underset{x}{\Σ}{\[\▿I\frac{\∂W}{\∂p}\]}^T\[T\left(x\right)-I\left(W\right(x;p\left)\right)\];$

Wherein H is hessian matrix, and W is the affine equation of burst, and T is alignment face, and I is actual picture, and x is in real image Pixel, p is the form parameter of shape in corresponding training process；

(2-2-8) form parameter p, wherein p=p+ Δ p are updated according to form parameter increment Delta p；It is then back to step (2-2-7), Continue to calculate Δ p by above-mentioned burst mapping function, until calculated form parameter increment Delta p is less than threshold value a or changes Generation number reaches maximum iteration time, then stop calculating；

(2-2-9) form parameter p obtaining after updating combines shape, and application principal component analytical method obtains the mesh followed the tracks of Mark facial image.

7. the video sequence face identification method based on AAM model according to claim 6 is it is characterised in that described threshold Value a is 500 to 2000.

8. the video sequence face identification method based on AAM model according to claim 6 is it is characterised in that described step Suddenly the posture trimming process in (2-2) is as follows：After accurately tracing into target facial image, obtain according to following the tracks of in iterative process Net shape parameter p posture correction is carried out to the subregion of face, the shape I (W (x after being reversed；P)) it is and positive face Consistent reconstruction face.

9. the video sequence face identification method based on AAM model according to claim 1 is it is characterised in that described step Suddenly obtaining characteristic vector η after dimensionality reduction in (2-3) is：

η=W_PCA ^Tu；

Optimal classification feature Γ of facial image to be identified is：

Γ=W_LDA ^TW_PCA ^Tu；

Wherein u is the second difference obtaining in step (2-3)；Wherein u=x-f；X is the posture correction obtaining in step (2-3) The picture after face subregion normalization afterwards, f is the average face obtaining in step (1-1).

10. the video sequence face identification method based on AAM model according to claim 1 is it is characterised in that described step Suddenly in (2-5) optimal classification feature Γ of facial image to be identified and training stage obtain each training picture optimal classification Feature Γ_ijMinimum euclid distance γ₁For：

${\mathrm{\γ}}_1=\underset{i,j}{min}\left|\right|\mathrm{\Γ}-{\mathrm{\Γ}}_{ij}\left|\right|,i=1,2,...,C,j=1,2,...N;$

Wherein C is the classification sum of the training picture after normalization, and N is the training picture sample sum that every apoplexy due to endogenous wind comprises.