CN101556699A - Face-based facial aging image synthesis method - Google Patents

Abstract

The invention discloses a face-based facial aging image synthesis method. The synthesis method comprises the following steps: (1) facial image data acquisition: establishing a facial database of different ages, obtaining a young facial image input by a user and obtaining various facial characteristic points by manually marking; (2) face matching based on individual algorithm: matching input images among different ages of the facial image database to find out a plurality of images by calculating a local curvature standard deviation of the facial characteristic points of face representation; (3) prototype synthesis by texture enhancement: utilizing the obtained images to perform the prototype synthesis by texture enhancement, and then obtaining two prototype an old person prototype image and young person prototype image; and (4) transformation of shapes and colors: performing transformation processing on the shapes and colors of the obtained two prototype images to obtain the final aging synthetic image. The synthesis method can help achieve realistic aging synthesis effects, such as wrinkle increase, eye bag generation, skin gloss dimming, hair whitening and the like.

Description

A kind of people's face aging image synthesis method based on shape of face

Technical field

The present invention relates to a kind of people's face aging image synthesis method based on shape of face.

Background technology

The aging simulation study on the synthesis of people's facial features looks all has important use to be worth in fields such as public security criminal detection, recognition of face, video display Cosmetic Design and public recreations.Criminal policeman and legal medical expert artists can be according to appearance photos then, according to experience and certain aging model, prediction escape convict or current appearance (the Age Progression of lost children for many years, http://www.forensicartist.com/agepro.html, AgeProgression, or infer to generate the appearance of criminal after the several years http://www.forensicartist.com/agepro.html); In production of film and TV, make up artists then utilize various makeup skills, and with reference to the older relatives' of performer appearance, generate the appearance effect of all ages and classes section for the performer makes up, thereby create the appearance of role's all ages and classes, and its age span can reach five, more than the threescore.

But the old and feeble forecast method of above-mentioned appearance mostly is based on experience, and manual synthetic method need spend a large amount of financial resources, and needs long professional training to grasp, and this has just limited popularizing of this technology.So far still there is not the old and feeble prediction of ripe quantification appearance generation system.

Utilizing the computer graphic image technology of recent development to come quantification ground prediction people face to change with the aging at age is one and is rich in challenging problem, its progress has guiding meaning for numerous areas such as criminal detection and video display cosmetic Aided Design, also is very interesting problem for teen-age Popular Science Education simultaneously.

Yet the influence of age for human face appearance is a suitable complex physical problem of mechanism, is subjected to all multifactor influences.Such as: individual's living environment, living and diet custom, job specification and degree of pressure etc., these factors all can be inscribed different years vestiges on a people's appearance.

In general, people changes in life appearance and is divided into two stages: the phase one is from childhood to youth, and the variation of this stage appearance mainly is the change of shape of face and face; Subordinate phase is from youth to old age, and this belongs to the gradual change of the different phase of growing up in stage, so its shape of face finalizes the design basically, and its appearance changes and is mainly reflected in the change of the colour of skin, hair, facial muscle relax level and color, glossiness.This paper research lays particular emphasis on the simulation to subordinate phase: i.e. the simulation of people's face aging course from the youth to the old stage.

Aspect the aging variation of people's face, existing so far many research work.Nineteen ninety-five, (Rowland D. such as Perrett, Perrett D.:Manipulating facial appearance through shape and color.IEEE Computer Graphics and Applications, 1995,15 (5): 70-76.) proposed a kind of prototyping, promptly realized that by changing shape and color the change of age of facial image is synthetic.(Tiddeman B. such as calendar year 2001 Tiddeman, Burt M., Perrett D.:Prototyping and transforming facialtextures for perception research.IEEE Computer Graphics and Applications, 2001,21 (5): 42-50) this method is expanded, and used method to realize that texture strengthens, and makes more remarkable with people's face variation effect of change of age based on wavelet filtering.But this method adopts same prototype to carry out aging to the somebody of institute to be changed, not to consider variety classes people's different old and feeble mode.

Lanitis etc. use the method based on the age function to realize the synthetic of people's face age transition, based on a facial image database that comprises 45 people at the photo of all ages and classes, utilize AAM (ActiveAppearance Models) method that everyone face is expressed as a proper vector, through statistical learning set up the age with the sign face characteristic vector between corresponding relation be the age function, the inverse function of obtaining each age function can realize obtaining the face characteristic vector and then facial image (the Lanitis A. at synthetic this age according to target age, Taylor C.J., Cootes T.F.:Toward automatic simulation of aging effectson face images.IEEE Trans.Pattern Analysis and Machine Intelligence, 2002,24 (4): 442-455.).But this method is mainly paid close attention to the appearance transition that the people extremely grew up from childhood, and is not suitable for old and feeble simulation the from the youth to old people's face.

Liu etc. have proposed a kind of surface details implantation technique (IBSDT) based on image, utilize this technology the grain details feature (as wrinkle, spot etc.) of old facial image can be transplanted on the young facial image, thereby aging effect (the Zicheng Liu of synthetic young facial image, Zhengyou Zhang, YingShan:Image-based surface detail transfer.IEEE Computer Graphics and Applications, 2004,24 (3): 30-35).(Zheng Nanning such as Zheng Nanning, pay sunlight, Zhang Ting, Zhuo Feng: the reconfiguration technique of the expression of people's face and age conversion and non-complete information (on). electronic letters, vol, 2003,31 (12A): 1955-1962.) proposed a kind of portrait agingization mapping algorithm based on old and feeble texture and otherness theory, the aging that utilizes the otherness gradual change technology of old and feeble texture scale map shadow casting technique and shape and texture to finish people's face changes.(Liu Jianyi such as Liu Jianyi, Zheng Nanning, ripple is bent in trip: method is combined in a kind of people's face aging based on small echo. the software journal, 2007 02 phases: 299-306.) proposed based on the old and feeble synthetic method of people's face of small echo, at first use small echo that picture breakdown is the high and low frequency part, again the low-frequency information of young facial image and the high-frequency information of old facial image are merged the old and feeble emulating image of generation people face.Often only with reference to the information and the template of one or two people's face, its old and feeble information and mode are abundant not enough in above-mentioned work.

Blanz and Vetter have set up faceform's database of being made up of 200 three-dimensional face models, deformation model (Morphable model) has been proposed, Model Matching by modeling of linear object class and Pixel-level, under Unified frame, realize the conversion (BlanzV. of attributes such as attitude, expression, age, illumination to people's face, Vetter T.:A morphable model for the synthesis of 3D faces.In ComputerGraphics Proc.SIGGRAPH ' 99,1999:187-194.).Scherbaum etc. expand above-mentioned model bank, the method of utilizing non-linear support vector to degenerate obtains the age function by study, thereby use the age transition trace simulation faceform's of nonlinear personalization age growth (Scherbaum K., Sunkel M., Seidel H.-P., Blanz V.:Prediction of Individual Non-Linear AgingTrajectories of Faces.Computer Graphics Forum, 2007,26 (3): 285-294.).The old and feeble high-frequency information (as wrinkle) of face that makes usually increases, (Golovinskiy A. such as Golovinskiy, MatusikW., Pfister H., Rusinkiewicz S., Funkhouser T.:A statistical model for synthesis ofdetailed facial geometry.In SIGGRAPH ' 06 2006:1025-1034.) makes it to mate the synthetic faceform's of method the old and feeble effect of faceform's statistic of target age by the local statistic of adjusting the input model migrated image.But these methods based on three-dimensional modeling then depend on the 3-D scanning portrait database of large sample, and its data volume is bigger, and operational processes is got up little convenient.

In sum, work in the past is the information based on single people's face usually; Perhaps handle proprietary aging, do not consider that the variety classes people has different personalized ageing approach by single-mode; Though external scholar has set up the old and feeble database of people's face of larger capacity, it only is suitable for the westerner, and does not fit into Asian's kind, and to still not deep enough to the aging research in old stage from the youth; And also exist data to obtain and unhandy shortcoming at present based on the old and feeble analogue technique of three-dimensional model, limited its application.

Invention Inner holds

The objective of the invention is to overcome the deficiencies in the prior art, a kind of people's face aging image synthesis method based on shape of face is provided.

People's face aging image synthesis method based on shape of face may further comprise the steps:

1) set up the face database of all ages and classes, obtain a young facial image of user's input simultaneously, manual markings obtains the unique point of everyone face;

2) local curvature's standard deviation of the unique point by calculating the shape of face feature that characterizes people's face is mated input picture and to be found out multiple image in all ages and classes section of facial image database;

3) use in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype figure pictures of old man's prototype and young man's prototype;

4) two prototype figures that obtain in input picture and the step 3) are looked like to carry out the conversion process of proterties and color, obtain the final aging composograph.

Described people sets up the face database of all ages and classes, obtains a young facial image of user's input simultaneously, obtains the unique point step of everyone face with existing algorithm:

(1) from the facial image file of all ages and classes section, the picture manual markings of obtaining young man's face of user's input is obtained the unique point of above-mentioned image;

Local curvature's standard deviation of described unique point by calculating the shape of face feature that characterizes people's face, input picture mated in all ages and classes section of facial image database find out the multiple image step:

(2) from the unique point that obtains, choose the left cheek of people's face bottom profiled, chin and 13 points of the right cheek (p ₁, p ₂... p ₁₃), it is carried out the match of cubic spline curve, on this curve, try to achieve the curvature (K of each unique point ₁, K ₂... K ₁₃);

(3) calculate the left cheek respectively, the standard deviation of the curvature value of the unique point at chin and right cheek place is used σ respectively ₁₁, σ ₂, σ ₁₂Expression, wherein σ ₁₁And σ ₁₂The projecting degree of expression people face two cheek the right and lefts, σ ₂The projecting degree at expression people face chin place uses σ ₁=(σ ₁₁+ σ ₁₂)/2 are as the tolerance of people's face two cheek place projecting degrees, v=(σ ₁, σ ₂) for the expression shape of face mellow and full/module of the degree of becoming thin;

(4) calculating input image is mellow and full/degree of becoming thin v _Input, and calculate v _InputMellow and full/the degree of becoming thin v with every width of cloth image i in the database _DbiEuclidean distance || v _Input-v _Dbi||, when || v _Input-v _Dbi|| be less than or equal to pre-set threshold t promptly || v _Input-v _Dbi|| during≤t, this storehouse image goes into to elect as the composing images of prototype of the age group at his place.

Described use is in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype image step of old man's prototype and young man's prototype:

(5) shape of each group prototype is the mean value of this group image shape, obtains by each the corresponding point coordinate in every width of cloth picture shape vector is asked on average:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{x}_i---\left(1\right)$

Wherein, x _iBe the shape vector of i width of cloth image, form by the x and the y coordinate figure of n unique point:

$x_i=(x_0^i,y_0^i,x_1^i,y_1^i,\·\·\·,x_n^i,y_n^i)---\left(2\right)$

In the formula: x represents the mean value of the shape vector of N width of cloth image, (x _j ⁱ, y _j ⁱ) be j the unique point coordinate that i opens facial image.

Prototype figure is that each width of cloth scalloping in the group is the mean value of each corresponding point after the average shape as each color of pixel value: at first use morphing will organize interior image and twist, for each corresponding pixel, ask for the mean value of color:

$\stackrel{\‾}{c}(x,y)=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{c}_i(W_x^i(x,y),W_y^i(x,y))---\left(3\right)$

Wherein, c _i(x is that expression i width of cloth image is at point (x, the vector of RGB color value y), (W y) _x ⁱ, W _y ⁱ) the distortion function of expression i width of cloth image, (x y) is N width of cloth image in point (x, average RGB color vector y) to c;

(6) select for use the real part branch of Gabor function as Hi-pass filter, cubic B-spline function is carried out following processing as low-pass filter again to each width of cloth picture construction image pyramid of forming prototype, on a wavelet sub-band w, carry out the low-pass filtering of x and y direction respectively, the level and smooth subband σ that obtains _wReflected the edge strength in the scope, for each point (x, y):

σ _w(x，y)＝h _x*h _y*|w(x，y)| (4)

H wherein _xAnd h _yBe one dimension cubic B-spline smoothing filter, * is the convolution symbol,

The mean value calculation of the level and smooth subband of N width of cloth image is as follows:

${\stackrel{\‾}{\mathrm{\σ}}}_w(x,y)=\frac{1}{N}\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{\mathrm{\σ}}_w^i(x,y)---\left(5\right)$

For a wavelet sub-band w, calculate the mean value w of N width of cloth image, the mean value σ behind it and the every width of cloth image smoothing is multiplied each other, remove smoothed image σ again with w _w, can obtain more representing N width of cloth picture edge characteristic value v (x, y):

$v(x,y)=\frac{\stackrel{\‾}{w}(x,y)\stackrel{\‾}{\mathrm{\σ}}(x,y)}{{\mathrm{\σ}}_{\stackrel{\‾}{w}}(x,y)}---\left(6\right)$

Replace w to carry out image reconstruction with v, obtain having the prototype that texture strengthens effect.

Described two prototype figures that obtain in input picture and the step 3) are looked like to carry out the conversion process of proterties and color, obtain final aging composograph step:

(7) right according to synthetic its personalized prototype of method mentioned above, carry out change of shape, the change of shape of image is expressed as follows:

x _os＝x _s+α×(x _op-x _yp) (7)

X wherein _sThe shape vector of expression input source image, x _OpAnd x _YpRepresent selected the elderly's face and young people's face prototype respectively, α is old and feeble ratio, x _OsBe target image and be the shape vector that source images is old and feeble after changing;

(8) source images and the right image of prototype all twisted be to carry out change color again by the target shape vector, calculate the color of target image for each pixel:

c _os(x，y)＝c _s(x，y)+α×(c _op(x，y)-c _yp(x，y)) (8)

(x, y) presentation video is at point (x, the vector of RGB color value y) for c wherein.

The present invention has personalization, has promptly guaranteed the similarity of old and feeble image and original young characteristics of image to greatest extent.Existing old and feeble algorithm is not often considered old and feeble personalized question, and the old and feeble effect of realization is stereotyped.We have proposed the personalized human face matching algorithm based on people's face outline local curvature standard deviation, have effectively weighed the similarity of two people's faces, thereby have realized old and feeble synthetic personalization.

This synthetic method can realize that wrinkle, pouch, skin gloss, hair turn the old and feeble variation effect of the sense of reality such as grey, can generate the old and feeble image of all ages and classes fast, and is easy and simple to handle, is easy to promote.And existing many old and feeble synthetic methods need the interactive operation of many complexity, use trouble.

In a word, using the present invention can fast and effeciently realize based on the old and feeble image of people's face of personalized prototype synthetic.The present invention has solved the lacking individuality that exists in the existing old and feeble synthetic method well, the deficiency of complicated operation, and on the convenience of the synthetic personalization of aging and user's operation, method of the present invention all is significantly increased.

Description of drawings

Fig. 1 is the face database synoptic diagram;

Fig. 2 is the human face characteristic point synoptic diagram;

Fig. 3 (a) is to be 25 years old input picture at the age;

Fig. 3 (b) is to be 30 years old output image at the age;

Fig. 3 (c) is to be 40 years old output image at the age;

Fig. 3 (d) is to be 50 years old output image at the age;

Fig. 3 (e) is to be 60 years old output image at the age;

Fig. 3 (f) is to be 70 years old output image at the age;

Fig. 4 (a) is to be 25 years old input picture at the age;

Fig. 4 (b) is to be 30 years old output image at the age;

Fig. 4 (c) is to be 40 years old output image at the age;

Fig. 4 (d) is to be 50 years old output image at the age;

Fig. 4 (e) is to be 60 years old output image at the age;

Fig. 4 (f) is to be 70 years old output image at the age.

Embodiment

People's face aging image synthesis method based on shape of face may further comprise the steps:

1) sets up the face database of all ages and classes, obtain a young facial image of user's input simultaneously, obtain the unique point of everyone face with existing algorithm;

2) local curvature's standard deviation of the unique point by calculating the shape of face feature that characterizes people's face is mated input picture and to be found out multiple image in all ages and classes section of facial image database;

3) use in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype figure pictures of old man's prototype and young man's prototype;

4) two prototype figures that obtain in input picture and the step 3) are looked like to carry out the conversion process of proterties and color, obtain the final aging composograph.

Described people sets up the face database of all ages and classes, obtains a young facial image of user's input simultaneously, obtains the unique point step of everyone face with existing algorithm:

(1) from the facial image file of all ages and classes section, the picture manual markings of obtaining young man's face of user's input is obtained the unique point of above-mentioned image

Local curvature's standard deviation of described unique point by calculating the shape of face feature that characterizes people's face, input picture mated in all ages and classes section of facial image database find out the multiple image step:

(2) from the unique point that obtains, choose the left cheek of people's face bottom profiled, chin and 13 points of the right cheek (p ₁, p ₂... p ₁₃), it is carried out the match of cubic spline curve, on this curve, try to achieve the curvature (K of each unique point ₁, K ₂... K ₁₃);

(3) calculate the left cheek respectively, the standard deviation of the curvature value of the unique point at chin and right cheek place is used σ respectively ₁₁, σ ₂, σ ₁₂Expression, wherein σ ₁₁And σ ₁₂The projecting degree of expression people face two cheek the right and lefts, σ ₂The projecting degree at expression people face chin place uses σ ₁=(σ ₁₁+ σ ₁₂)/2 are as the tolerance of people's face two cheek place projecting degrees, v=(σ ₁, σ ₂) for the expression shape of face mellow and full/module of the degree of becoming thin,

V=in this example (0.012,0.023);

(4) calculating input image is mellow and full/degree of becoming thin v _Input, and calculate v _InputMellow and full/the degree of becoming thin v with every width of cloth image i in the database _DbiEuclidean distance || v _Input-v _Dbi||, when || v _Input-v _Dbi|| be less than or equal to pre-set threshold t promptly || v _Input-v _Dbi|| during≤t, this storehouse image goes into to elect as the composing images of prototype of the age group at his place, t=0.005 in this example, and we have obtained 28 of the young facial images that are complementary with input picture, 32 of old man's facial images.

Described use is in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype image step of old man's prototype and young man's prototype:

(5) shape of each group prototype is the mean value of this group image shape, obtains by each the corresponding point coordinate in every width of cloth picture shape vector is asked on average:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{x}_i---\left(1\right)$

Wherein, x _iBe the shape vector of i width of cloth image, form by the x and the y coordinate figure of n unique point:

$x_i=(x_0^i,y_0^i,x_1^i,y_1^i,\·\·\·,x_n^i,y_n^i)---\left(2\right)$

In the formula: x represents the mean value of the shape vector of N width of cloth image, (x _j ⁱ, y _j ⁱ) be j the unique point coordinate that i opens facial image.

Prototype figure is that each width of cloth scalloping in the group is the mean value of each corresponding point after the average shape as each color of pixel value: at first use morphing will organize interior image and twist, for each corresponding pixel, ask for the mean value of color:

$\stackrel{\‾}{c}(x,y)=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{c}_i(W_x^i(x,y),W_y^i(x,y))---\left(3\right)$

Wherein, c _i(x is that expression i width of cloth image is at point (x, the vector of RGB color value y), (W y) _x ⁱ, W _y ⁱ) the distortion function of expression i width of cloth image, (x y) is N width of cloth image in point (x, average RGB color vector y) to c;

(6) select for use the real part branch of Gabor function as Hi-pass filter, cubic B-spline function is carried out following processing as low-pass filter again to each width of cloth picture construction image pyramid of forming prototype, on a wavelet sub-band w, carry out the low-pass filtering of x and y direction respectively, the level and smooth subband σ that obtains _wReflected the edge strength in the scope, for each point (x, y):

σ _w(x，y)＝h _x*h _y*|w(x，y)| (4)

H wherein _xAnd h _yBe one dimension cubic B-spline smoothing filter, * is the convolution symbol,

The mean value calculation of the level and smooth subband of N width of cloth image is as follows:

${\stackrel{\‾}{\mathrm{\σ}}}_w(x,y)=\frac{1}{N}\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{\mathrm{\σ}}_w^i(x,y)---\left(5\right)$

For a wavelet sub-band w, calculate the mean value w of N width of cloth image, the mean value σ behind it and the every width of cloth image smoothing is multiplied each other, remove smoothed image σ again with w _w, can obtain more representing N width of cloth picture edge characteristic value v (x, y):

$v(x,y)=\frac{\stackrel{\‾}{w}(x,y)\stackrel{\‾}{\mathrm{\σ}}(x,y)}{{\mathrm{\σ}}_{\stackrel{\‾}{w}}(x,y)}---\left(6\right)$

Replace w to carry out image reconstruction with v, obtain having the prototype that texture strengthens effect.

Described two prototype figures that obtain in input picture and the step 3) are looked like to carry out the conversion process of proterties and color, obtain final aging composograph step:

(7) right according to synthetic its personalized prototype of method mentioned above, carry out change of shape, the change of shape of image is expressed as follows:

x _os＝x _s+α×(x _op-x _yp) (7)

X wherein _sThe shape vector of expression input source image, x _OpAnd x _YpRepresent selected the elderly's face and young people's face prototype respectively, α is old and feeble ratio, x _OsBe target image and be the shape vector that source images is old and feeble after changing; In this example, α=0.4 representative increases by 10 years old, and α=0.6 representative increases by 20 years old, and α=0.8 representative increases by 30 years old, and α=1.0 representatives increase by 40 years old, and α=1.2 representatives increase by 50 years old.

(8) source images and the right image of prototype all twisted be to carry out change color again by the target shape vector, calculate the color of target image for each pixel:

c _os(x，y)＝c _s(x，y)+α×(c _op(x，y)-c _yp(x，y))(8)

(x, y) presentation video is at point (x, the vector of RGB color value y) for c wherein.

By above step, can everybody realize any youth based on the synthetic face of the old and feeble image of people's face of personalized prototype.

What more than enumerate only is specific embodiments of the invention.Obviously, the invention is not restricted to above embodiment, many distortion can also be arranged.All distortion that those of ordinary skill in the art can directly derive or associate from content disclosed by the invention all should be thought protection scope of the present invention.

Claims (5)

1, a kind of people's face aging image synthesis method based on shape of face is characterized in that may further comprise the steps:

1) set up the face database of all ages and classes, obtain a young facial image of user's input simultaneously, manual markings obtains the unique point of everyone face;

2) local curvature's standard deviation of the unique point by calculating the shape of face feature that characterizes people's face is mated input picture and to be found out multiple image in all ages and classes section of facial image database;

3) use in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype figure pictures of old man's prototype and young man's prototype;

4) two prototype figures that obtain in input picture and the step 3) are looked like to carry out the conversion process of proterties and color, obtain the final aging composograph.

2, a kind of people's face aging image synthesis method according to claim 1 based on shape of face, it is characterized in that described people sets up the face database of all ages and classes, obtain a young facial image of user's input simultaneously, obtain the unique point step of everyone face with existing algorithm:

(1) from the facial image file of all ages and classes section, the picture manual markings of obtaining young man's face of user's input is obtained the unique point of above-mentioned image.

3, a kind of people's face aging image synthesis method according to claim 1 based on personalized prototype, the local curvature's standard deviation that it is characterized in that described unique point by calculating the shape of face feature that characterizes people's face, input picture mated in all ages and classes section of facial image database find out the multiple image step:

(2) from the unique point that obtains, choose the left cheek of people's face bottom profiled, chin and 13 points of the right cheek (p ₁, p ₂... p ₁₃), it is carried out the match of cubic spline curve, on this curve, try to achieve the curvature (K of each unique point ₁, K ₂... K ₁₃);

(3) calculate the left cheek respectively, the standard deviation of the curvature value of the unique point at chin and right cheek place is used σ respectively ₁₁, σ ₂, σ ₁₂Expression, wherein σ ₁₁And σ ₁₂The projecting degree of expression people face two cheek the right and lefts, σ ₂The projecting degree at expression people face chin place uses σ ₁=(σ ₁₁+ σ ₁₂)/2 are as the tolerance of people's face two cheek place projecting degrees, v=(σ ₁, σ ₂) for the expression shape of face mellow and full/module of the degree of becoming thin;

(4) calculating input image is mellow and full/degree of becoming thin v _Input, and calculate v _InputMellow and full/the degree of becoming thin v with every width of cloth image i in the database _DbiEuclidean distance ‖ v _Input-v _Dbi‖ is as ‖ v _Input-v _DbiIt is ‖ v that ‖ is less than or equal to pre-set threshold t _Input-v _Dbi|| during≤t, this storehouse image goes into to elect as the composing images of prototype of the age group at his place.

4, a kind of people's face aging image synthesis method according to claim 1 based on personalized prototype, it is characterized in that described use is in step 2) in the image that finds to carry out the prototype that texture strengthens synthetic, obtain two prototype image step of old man's prototype and young man's prototype:

(5) shape of each group prototype is the mean value of this group image shape, obtains by each the corresponding point coordinate in every width of cloth picture shape vector is asked on average:

$\stackrel{\‾}{x}=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{x}_i---\left(1\right)$

Wherein, x _iBe the shape vector of i width of cloth image, form by the x and the y coordinate figure of n unique point:

$x_i=(x_0^i,y_0^i,x_1^i,y_1^i,...,x_n^i,y_n^i)---\left(2\right)$

In the formula: x represents the mean value of the shape vector of N width of cloth image, (x _j ⁱ, y _j ⁱ) be j the unique point coordinate that i opens facial image,

Prototype figure is that each width of cloth scalloping in the group is the mean value of each corresponding point after the average shape as each color of pixel value: at first use morphing will organize interior image and twist, for each corresponding pixel, ask for the mean value of color:

$\stackrel{\‾}{c}(x,y)=\frac{1}{N}\underset{i=0}{\overset{N}{\mathrm{\Σ}}}{c}_i(W_x^i(x,y),W_y^i(x,y))---\left(3\right)$

Wherein, c _i(x is that expression i width of cloth image is at point (x, the vector of RGB color value y), (W y) _x ⁱ, W _y ⁱ) the distortion function of expression i width of cloth image, (x y) is N width of cloth image in point (x, average RGB color vector y) to c;

(6) select for use the real part branch of Gabor function as Hi-pass filter, cubic B-spline function is carried out following processing as low-pass filter again to each width of cloth picture construction image pyramid of forming prototype, on a wavelet sub-band w, carry out the low-pass filtering of x and y direction respectively, the level and smooth subband σ that obtains _wReflected the edge strength in the scope, for each point (x, y):

σ _w(x，y)＝h _x*h _y*|w(x，y)| (4)

H wherein _xAnd h _yBe one dimension cubic B-spline smoothing filter, * is the convolution symbol,

The mean value calculation of the level and smooth subband of N width of cloth image is as follows:

${\stackrel{\‾}{\mathrm{\σ}}}_w(x,y)=\frac{1}{N}\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{\mathrm{\σ}}_w^i(x,y)---\left(5\right)$

For a wavelet sub-band w, calculate the mean value w of N width of cloth image, the mean value σ behind it and the every width of cloth image smoothing is multiplied each other, remove smoothed image σ again with w _w, can obtain more representing N width of cloth picture edge characteristic value v (x, y):

$v(x,y)=\frac{\stackrel{\‾}{w}(x,y)\stackrel{\‾}{\mathrm{\σ}}(x,y)}{{\mathrm{\σ}}_{\stackrel{\‾}{w}}(x,y)}---\left(6\right)$

Replace w to carry out image reconstruction with v, obtain having the prototype that texture strengthens effect.

5, a kind of people's face aging image synthesis method according to claim 1 based on personalized prototype, it is characterized in that described two prototype figures that obtain in input picture and the step 3) being looked like to carry out the conversion process of proterties and color, obtain final aging composograph step:

(7) right according to synthetic its personalized prototype of method mentioned above, carry out change of shape, the change of shape of image is expressed as follows:

x _os＝x _s+α×(x _op-x _yp) (7)

X wherein _sThe shape vector of expression input source image, x _OpAnd x _YpRepresent selected the elderly's face and young people's face prototype respectively, α is old and feeble ratio, x _OsBe target image and be the shape vector that source images is old and feeble after changing;

(8) source images and the right image of prototype all twisted be to carry out change color again by the target shape vector, calculate the color of target image for each pixel:

c _os(x，y)＝c _s(x，y)+α×(c _op(x，y)-c _yp(x，y)) (8)

(x, y) presentation video is at point (x, the vector of RGB color value y) for c wherein.

Images