CN104732200A - Skin type and skin problem recognition method - Google Patents

Abstract

The invention relates to the technical field of face recognition, in particular to a skin type and skin problem recognition method. The skin type and skin problem recognition method sequentially comprise the following steps of obtaining a face image, detecting whether the face image includes a face or not, detecting the color and luster and chrominance of the skin, calculating a texture contrast ratio value, calculating a mean gray value, matching a calculation result with a preset database and outputting a skin problem result obtained through matching. By the adoption of the skin type and skin problem recognition method, a computer can have the capacity of analyzing the skin type and skin problem of an object in the face image, achieve automatic intelligent management, analyze the skin of the object so as to obtain a plurality of indexes including skin color, skin blotches, moisture, oil, texture, wrinkles, skin pores and the like and perform comprehensive evaluation on the skin surface condition.

Description

The recognition methods of a kind of skin type and skin problem

Technical field

The present invention relates to technical field of face recognition, particularly the recognition methods of a kind of skin type and skin problem.

Background technology

Face is one of important biological characteristic, facial image has contained a large amount of information, such as sex, age, ethnic group, identity etc.The development of human-computer interaction technology (HCI) makes computer vision, artificial intelligence, plays an increasingly important role in monitoring, GUI Human Machine Interface etc.Along with the progress of technology, also become the focus of Recent study based on facial image pattern recognition problem.Comprising all kinds of identification problems such as Face datection, face identification, face character (sex, age, expression, race etc.) identifications.

Along with the development of modern society, the requirement of people to autologous skin health is also more and more higher.Facility helps user to find oneself skin type and skin problem just to seem more meaningful rapidly.The present invention can gather user's human face photo by mobile device terminal, is analyzed by intelligent skin recognition system comparison film, and by result feedback to user.

Chinese patent application CN 102831408A discloses a kind of face identification method, and it comprises the following steps: S1 judges to detect the certain gestures that whether there is user's setting in video sequence; If so, S2 is carried out; Otherwise, repeat step S1; S2 detects in video area whether have user's face, if so, carries out step S3, otherwise repeats step S2; S3 carries out real-time follow-up to the user's face detected; S4 extracts illumination-insensitive feature to user's facial image; S5 analog subscriber postural change rule; S6 face recognition process, judge whether the sample in facial image to be identified and face database belongs to a people together, if so, identifying terminates, and system performs user operation; If not, step S7 is carried out; S7 inquires whether user is first time use system; If so, newly-built Sample Storehouse, if not, whether inquiry user substitutes Sample Storehouse, if so, this user's facial image is replaced the original sample in face database, otherwise end operation.

From prior art, can also recognize: Chinese patent application CN 103729625 A discloses a kind of method of recognition of face, and Chinese patent application CN 103761504 A discloses a kind of face identification system.

But, above-mentioned prior art only can provide a kind of system or method of recognition of face, face image data can not be utilized further, more detailed analyzing and processing is provided, such as to the analytic function of the sex of user, skin, face characteristic similarity and age etc., to cater to the particular demands of user.

Summary of the invention

For overcoming the deficiencies in the prior art, the invention provides the recognition methods of a kind of skin type and skin problem, its special feature is to comprise the following steps successively:

S1, use camera head obtain facial image;

S2, detect this facial image and whether comprise face, and when determining that this facial image comprises face, jump procedure S3;

S3, carry out skin colour and colorimetric detection to this facial image, the computing formula of wherein this detection is:

${\mathrm{\μ}}_c=\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{p}_{\mathrm{ij}}^c$

${\mathrm{\σ}}_c={\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^2\right]}^{1/2}$

${{\mathrm{\θ}}_c=\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^3\right]}^{1/3}$

Wherein, M, N are the two-dimensional of image, and i, j represent the row, column of the place image of this pixel respectively, and c is the value of the color component of described pixel;

S4, define texture co-occurrence matrix C (i, j), and utilize following formulae discovery to obtain texture comparison's angle value:

$\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}{(i-j)}^{2}C(i,j)\text{;}$

S5, utilize difference and the gray-scale value Fast Match Algorithm of algorithm and this facial image, obtain following computing formula:

$R(i,j)=\frac{\underset{m,n}{\mathrm{\Σ}}|S^{i,j}(m,n)-E\left(S^{i,j}\right)||T(m,n)-E\left(T\right)|}{\sqrt{\underset{m,n}{\mathrm{\Σ}}{(S^{i,j}(m,n)-E\left(S^{i,j}\right))}^2\underset{m,n}{\mathrm{\Σ}}{(T(m,n)-E\left(T\right))}^2}}$

Wherein, E (S ^i,j) with E (T) be user's skin subgraph S respectively ^i,jwith the average gray of the skin image T in presetting database;

S6, abovementioned steps S3 to S5 is calculated acquired results mate with this pre-set skin issue database, and the skin problem result of output matching gained.

According to one embodiment of present invention, before this recognition methods of enforcement, the skin image storehouse comprising various typical face is first set up.

According to one embodiment of present invention, this step S2 utilizes network neural method, character references method or primary standard of color method whether to comprise face to detect this facial image.

According to one embodiment of present invention, before this step S3 of execution, perform photo-irradiation treatment to described facial image, described facial image is utilized histogram specification conversion, and the computing formula of wherein said conversion is:

$g(x,y)=a+\frac{\ln \left(f(x,y)\right)+1}{b\ln c}.$

Wherein, f (x, y) is the pixel element data of input facial image, and g (x, y) is the histogrammic pixel element data of output gray level, the parameter that a, b, c are the adjustment position of curve and shape and introduce.Further, after utilizing histogram specification to change described facial image, then carry out 3 × 3 medium filterings, remove the noise information introduced in image.

By technique scheme, the recognition methods according to skin type of the present invention and skin problem can realize following technique effect:

(1) the subject's skin type in imparting Computer Analysis facial image and the ability of skin problem;

(2) automatic intelligent management is realized

(3) subject's skin is analyzed, draw some indexs such as the colour of skin, spot, moisture oil content, texture wrinkle and pore, comprehensive evaluation is carried out to skin surface situation.

Accompanying drawing explanation

Fig. 1 shows the method flow diagram of identification skin type according to the embodiment of the present invention and skin problem.

Fig. 2 shows the other method process flow diagram of identification skin type according to the embodiment of the present invention and skin problem.

Embodiment

Describe the present invention in detail below in conjunction with accompanying drawing and specific implementation method, be used for explaining the present invention in exemplary embodiment and description of the present invention, but not as a limitation of the invention.

Before implementing the recognition methods according to the embodiment of the present invention, the skin image storehouse comprising various typical face can be set up in advance.Research and the experiment of current most of face aspect problem are all carried out based on one or several facial image databases.In order to realize the research of face skin type and skin problem recognizer and practical, the universal standard facial image database of the recognizer for realizing face skin type and skin problem can be set up, image in this image library has certain restriction on shooting condition, size for example for face must constrain in the scope of certain regulation, the Gray Face image of different illumination conditions, and strictly having divided training set, storehouse (Gallery), different test set etc., this is one of most widely used face database of field of face identification.

As shown in Figure 1, after obtaining facial image by camera head, first pre-service is carried out to obtained image, comprise and confirm face, facial pre-service.

In the present embodiment, confirm that the method for face can comprise: the method for network neural method (Neural Network), character references method (Feature-based) and primary standard of color method (Color-based).Wherein, network neural method uses many training set data (Training Data) to go training network neural, and these data somes are face, a part of Shi Fei face, system picked out comprise the image of face.Character references method utilizes some features of people face to detect face, and such as: the eyes on face, nose, face, and these organs have fixing relative position; And with regard to whole face, the chances are ovalize, and between face and background, have edge line roughly, etc., these features all can go out face for systems scan.And primary standard of color rule utilizes the color of people face to judge whether certain image comprises face: well-known, the color of face generally includes the colors such as such as orange, yellow, white, brown, dark brown; If image is GTG, the change of the GTG color of face can't be too large, and eyes, face, hair are more black parts; Thus, according to the feature of the organs such as eyes, eyebrow, face, nose and geometry site each other, face can be detected.

Detect in this image after comprising face, usual needs carry out pre-service to this image, this is because often there is the even problem of uneven illumination in pending facial image, and this directly can have influence on the feature extraction precision of face, therefore photo-irradiation treatment must be carried out to improve picture quality to the image of input.Usually, grey level histogram can be used for the statistical relationship in representative digit image between each gray level and frequency of its appearance.For partially secretly, the image that partially bright, brightness range is not enough or contrast is not enough carries out histogram specification, and the histogram distribution of input picture can be made to be transformed into approximate specific histogram.Transforming function transformation function can select the function of the forms such as such as Gauss, Rayleigh, logarithm, index.Adopt log-transformation form in the present embodiment:

$g(x,y)=a+\frac{\ln \left(f(x,y)\right)+1}{b\ln c}.$

In formula, f (x, y) is input picture, and g (x, y) is output image, the parameter that a, b, c are the adjustment position of curve and shape and introduce.The low tonal range of image can be made to be expanded by this conversion, high tonal range can be compressed, and the intensity profile of image is tending towards even.Preferably, the image after greyscale transformation is carried out to the medium filtering of such as 3 × 3 again, remove the noise information introduced in image, to improve picture quality.

After extracting face characteristic, use according to the method for the embodiment of the present invention to obtain Output rusults needed for required skin type/problem, comprise: detection of skin color and luster with colourity, calculate texture comparison's angle value, calculate average gray and above-mentioned calculating acquired results is mated with this presetting database, and the skin problem result of output matching gained.

In the present embodiment, so a kind of algorithm can be used with detection of skin color and luster and colourity, wherein, characterize a colorized face images with the colour of skin color matrix/color intensity distribution sign.Most color distribution information can be represented by three second-order matrix, wherein, what one line rank matrix (μ c) characterized is average color, and what rank, two wires matrix (δ c) characterized is standard deviation, and third moment sign is measure of skewness (θ c) color.Utilize following mathematical formulae, extract each three kinds of look planes (R, G, B) from these three low submatrixs (μ c, δ c, θ C):

${\mathrm{\μ}}_c=\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{p}_{\mathrm{ij}}^c$

${\mathrm{\σ}}_c={\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^2\right]}^{1/2}$ ${{\mathrm{\θ}}_c=\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^3\right]}^{1/3}$

Wherein, M, N are the two-dimensional of image, and i, j represent the place row, column of this pixel respectively, and c is the value of color component.Consequently, only need extraction nine parameters as the feature of colorized face images, such as comprise look plane parameter (RGB), average color, standard deviation, measure of skewness color, color and luster chromatic value, texture contrast, average gray etc., by the calculating to above skin colour and colourity, the series of parameters arrays such as the input subject's skin color and luster of facial image and colourity can be drawn.

In the present embodiment, texture comparison's angle value is calculated with skin-texture detection algorithm.Texture is the feature of facial image, and one of texture itself large feature is the repetition of its pattern.At this, introduce term " texture primitive ", namely its meaning refers to the mode unit of texture, and the size of texture primitive, shape, color and orientation can change in very large interval, and the difference between any two textures can be presented as the intensity of variation of texture primitive.Can build a texture co-occurrence matrix, this texture co-occurrence matrix C (i, j) is by displacement vector D _x, D _y=(Δ X, Δ Y) defines, and wherein Δ X, Δ Y are the displacements respectively on x direction and y direction, then calculates all pixels and to be separated by displacement D _x, D _ythe gray level i had and j.It also may be because the spatial statistics of the texture primitive in image distributes, and comprises the important information such as structural arrangement about the imaging surface in bottom.Then, each element in Matrix C (I, j) is normalized.By to above calculating, can calculate the series of parameters array of dermatoglyph, the computing formula of texture contrast numerical value is:

Can obtain thus inputting subject's skin texture in facial image to degree than size, also just can represent the weight of subject's skin texture.

In the present embodiment, utilize the gray-scale value Fast Match Algorithm of difference and algorithm and this facial image, obtain following computing formula:

$R(i,j)=\frac{\underset{m,n}{\mathrm{\Σ}}|S^{i,j}(m,n)-E\left(S^{i,j}\right)||T(m,n)-E\left(T\right)|}{\sqrt{\underset{m,n}{\mathrm{\Σ}}{(S^{i,j}(m,n)-E\left(S^{i,j}\right))}^2\underset{m,n}{\mathrm{\Σ}}{(T(m,n)-E\left(T\right))}^2}}.$

Wherein, E (S ^i,j) with E (T) be user's skin subgraph S respectively ^i,jwith the average gray of the skin image T (m, n) in presetting database.

Then, by extracting the numerical value of above gained, carrying out Data Matching with the skin problem database preset, finally obtaining the skin problem mated, and being back to client.

Above the technical scheme that the embodiment of the present invention provides is described in detail, apply specific case herein to set forth the principle of the embodiment of the present invention and embodiment, the explanation of above embodiment is only applicable to the principle helping to understand the embodiment of the present invention; Meanwhile, for one of ordinary skill in the art, according to the embodiment of the present invention, embodiment and range of application all will change, and in sum, this description should not be construed as limitation of the present invention.

Claims (5)

1. a recognition methods for skin type and skin problem, is characterized in that, comprises the following steps successively:

S1, use camera head obtain facial image;

S2, detect described facial image and whether comprise face, and when determining that described facial image comprises face, jump procedure S3;

S3, carry out skin colour and colorimetric detection to described facial image, the computing formula of wherein said detection is:

${\mathrm{\μ}}_c=\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{p}_{\mathrm{ij}}^c$

${\mathrm{\σ}}_c={\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^2\right]}^{1/2}$

${\mathrm{\θ}}_c={\left[\frac{1}{\mathrm{MN}}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{(p_{\mathrm{ij}}^c-{\mathrm{\μ}}_c)}^3\right]}^{1/3}$

Wherein, M, N are the two-dimensional of image, and i, j represent the row, column of the place image of described pixel respectively, and c is the value of the color component of described pixel;

S4, define texture co-occurrence matrix C (i, j), and utilize following formulae discovery to obtain texture comparison's angle value:

$\underset{i}{\mathrm{\Σ}}\underset{j}{\mathrm{\Σ}}{(i-j)}^{2}C(i,j);$

S5, utilize difference and the gray-scale value Fast Match Algorithm of algorithm and described facial image, obtain following computing formula:

$R(i,j)=\frac{\underset{m,n}{\mathrm{\Σ}}|S^{i,j}(m,n)-E\left(S^{i,j}\right)||T(m,n)-E\left(T\right)|}{\sqrt{\underset{m,n}{\mathrm{\Σ}}(S^{i,j}(m,n)-E\left(S^{i,j}\right){)}^2\underset{m,n}{\mathrm{\Σ}}{(T(m,n)-E\left(T\right))}^2}}$

Wherein, E (S ^i,j) with E (T) be user's skin subgraph S respectively ^i,jwith the average gray of the skin image T in presetting database;

S6, abovementioned steps S3 to S5 is calculated acquired results mate with described pre-set skin issue database, and the skin problem result of output matching gained.

2. recognition methods according to claim 1, is characterized in that, before the described recognition methods of enforcement, first sets up the skin image storehouse comprising various typical face.

3. recognition methods according to claim 1, is characterized in that, described step S2 utilizes network neural method, character references method or primary standard of color method whether to comprise face to detect described facial image.

4. recognition methods according to claim 1, is characterized in that, before the described step S3 of execution, perform photo-irradiation treatment to described facial image, described facial image is utilized histogram specification conversion, and the computing formula of wherein said conversion is:

$g(x,y)=a+\frac{\ln \left(f(x,y)\right)+1}{b\ln c}.$

Wherein, f (x, y) is the pixel element data of input facial image, and g (x, y) exports the pixel element data of facial image, and a, b, c are the parameters adjusting the position of curve and shape and introduce.

5. recognition methods according to claim 4, is characterized in that, after being utilized by described facial image histogram specification to change, then carries out 3 × 3 medium filterings, removes the noise information introduced in image.