Face aging prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a face aging prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: receiving an input face image, a real age, a predicted age and a maintenance frequency; analyzing a real skin state based on the face image; calculating an aging factor based on the real skin state; calculating a maintenance factor based on the maintenance frequency; an aging coefficient is calculated based on the real age, the predicted age, the aging factor, and the care factor. According to the method, network training and collection of face data for network training are not needed, and the aging coefficient related to the predicted age can be rapidly predicted by extracting the skin features according to the face image corresponding to the real age of the user and combining the maintenance frequency, namely, the face aging speed of the user is rapidly predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of skin detection, and in particular to a method, device, electronic equipment and storage medium for predicting facial aging. Background Art

[0002] As people age, their facial features will show a certain degree of aging. Bad living habits and lack of maintenance may accelerate the aging of the face. The aging prediction algorithm can predict the aging of facial images, allowing people to intuitively observe the characteristics of the face after aging.

[0003] The existing face aging prediction algorithms are directly predicted through the image generation network, and the training of the image generation network depends on a large amount of data; and they are not based on the feature evolution of the user's current facial features, lacking differentiation. In addition, the existing face aging prediction algorithms are only based on face images collected under white light, that is, white light images. The features that can be seen in white light images are limited, and some deep features can only be observed with the help of other light sources. The results predicted based on white light images are also difficult to fit the user's aging trend. Therefore, a set of aging prediction methods based on multiple light sources and strong correlation with user features are needed to perform aging prediction. Summary of the invention

[0004] The main purpose of the present invention is to propose a face aging prediction method, device, electronic device and storage medium, which overcome the shortcomings of the prior art and do not require network training or collection of face data for network training. It only needs to extract skin features based on the face image corresponding to the user's real age, and combine the maintenance frequency to quickly predict the aging coefficient related to the predicted age, that is, quickly predict the user's face aging speed.

[0005] The present invention adopts the following technical solution:

[0006] In a first aspect, a method for predicting facial aging includes:

[0007] Receiving an input face image, real age, predicted age and maintenance frequency; the input image includes at least one;

[0008] Analyze the real skin condition based on the face image;

[0009] An aging factor is calculated based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition;

[0010] Calculate a maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, hydration frequency, and staying up late frequency, and the maintenance factor is inversely correlated with the sun protection frequency and the hydration frequency, and is positively correlated with the staying up late frequency;

[0011] The aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor; among them, the aging coefficient is positively correlated with the aging factor, maintenance factor and the difference between the predicted age and the real age.

[0012] Preferably, the aging coefficient is calculated based on the real age, the predicted age, the aging factor and the maintenance factor, and is expressed as follows:

[0013] F=a^(z-x+b)

[0014] Among them, F represents the aging coefficient; x represents the real age; z represents the predicted age; a represents the aging factor; and b represents the maintenance factor.

[0015] Preferably, the aging factor is calculated based on the actual skin condition, as shown below:

[0016] a=a1+a2*(1+c)

[0017] Among them, a represents the aging factor; a1 and a2 are constants; and c represents the real skin condition coefficient.

[0018] Preferably, the real skin condition is analyzed based on the face image, which is expressed as follows:

[0019]

[0020] Among them, c represents the real skin condition coefficient; x represents the real age; fn represents the score of the skin features extracted based on the face image, increasing from 0 to 1; N represents the number of skin features.

[0021] Preferably, the skin characteristics include at least one of white spots, wrinkles, brown spots, UV spots and pigmentation.

[0022] Preferably, the white light spots are acquired based on facial images collected under white light; the wrinkles are acquired based on facial images collected under cross-polarized light; the brown spots are acquired based on facial images collected under parallel polarized light; the ultraviolet spots are acquired based on facial images collected under ultraviolet light; and the pigmentation is acquired based on facial images collected under blue light.

[0023] Preferably, the calculation method of the score of skin characteristics includes:

[0024] Extract the mask corresponding to the skin features from the face image;

[0025] Calculate the ratio of the area of ​​the mask of the skin feature to the facial skin area in the facial image;

[0026] The proportion is used as a score of the skin feature.

[0027] Preferably, the maintenance factor is calculated based on the maintenance frequency, as shown below:

[0028]

[0029] Among them, b represents the maintenance factor; b1 and b1 are constants; V represents the frequency of sun protection; W represents the frequency of hydration; and T represents the frequency of staying up late.

[0030] Preferably, after calculating the aging coefficient based on the real age, the predicted age, the aging factor and the maintenance factor, the method further includes:

[0031] Predicting the skin characteristics of the predicted age according to the aging coefficient to obtain predicted skin characteristics; the predicted skin characteristics at least include spots and wrinkles;

[0032] The predicted skin features are displayed on the white light face image.

[0033] Preferably, the skin characteristics of the predicted age are predicted according to the aging coefficient to obtain the predicted skin characteristics, which specifically includes:

[0034] Extracting masks of skin features related to predicted skin features from face images acquired under different light sources;

[0035] Adding and / or superimposing masks of relevant skin features to obtain a predicted skin feature mask;

[0036] A convolution operation is performed on the predicted feature mask and a convolution kernel related to the aging coefficient, and channel processing is performed based on the predicted feature mask after the convolution operation to obtain the predicted skin features.

[0037] In a second aspect, a face aging prediction device includes:

[0038] An input data receiving module, used to receive an input face image, real age, predicted age and maintenance frequency; the input image includes at least one;

[0039] A real skin condition acquisition module is used to analyze the real skin condition based on the face image;

[0040] An aging factor calculation module, used to calculate the aging factor based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition;

[0041] A maintenance factor calculation module, used to calculate the maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, water replenishment frequency and staying up late frequency, and the maintenance factor is inversely related to the sun protection frequency and the water replenishment frequency, and is positively related to the staying up late frequency;

[0042] The aging coefficient calculation module is used to calculate the aging coefficient based on the real age, predicted age, aging factor and maintenance factor; wherein the aging coefficient is positively correlated with the aging factor, maintenance factor and the difference between the predicted age and the real age.

[0043] Furthermore, the face aging prediction device further includes:

[0044] A skin feature prediction module, used to predict the skin features of the predicted age according to the aging coefficient to obtain predicted skin features; the predicted skin features at least include spots and wrinkles;

[0045] The display module is used to display the predicted skin features on the white light face image.

[0046] In a third aspect, an electronic device is provided, characterized in that it comprises a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to execute the steps in the face aging prediction method.

[0047] In a fourth aspect, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for a processor to load to execute the steps of the face aging prediction method.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] (1) The present invention does not require network training or the collection of face data for network training. It only needs to extract skin features based on the face image corresponding to the user's real age, and can quickly predict the aging coefficient related to the predicted age by combining the maintenance frequency, that is, quickly predict the user's face aging speed;

[0050] (2) When extracting skin features based on face images, the present invention uses face images under different light sources to extract different skin features, including extracting white light spots based on white light, extracting wrinkles based on cross-polarized light, extracting brown spots based on parallel polarized light, extracting ultraviolet spots based on ultraviolet light, extracting pigmentation based on blue light, etc., thereby being able to extract a lot of feature information that cannot be obtained under white light, so that aging prediction can be more in line with the user's aging trend;

[0051] (3) The present invention performs feature aging prediction of various dimensions on white light face images based on facial skin feature representation of multi-light source face images, and fuses them into final aging prediction results; the feature analysis results and aging prediction results are saved to track the changing trends of facial features and aging prediction results.

[0052] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention so that it can be implemented in accordance with the contents of the specification, and in order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below.

[0053] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of a method for predicting facial aging according to an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the relationship between the aging coefficient and age according to an embodiment of the present invention;

[0056] Figure 3 A flowchart of displaying predicted skin features in a white light face image according to an embodiment of the present invention;

[0057] Figure 4 is a structural block diagram of a face aging device according to an embodiment of the present invention;

[0058] Figure 5 It is a structural block diagram of predicting skin features in a white light face image according to an embodiment of the present invention;

[0059] Figure 6 4 is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments, and all other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making creative work are within the scope of protection of the present invention.

[0061] In the description of the present invention, it should be noted that the terms "comprises", "includes" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0062] In the description of the present invention, it should be noted that, unless otherwise clearly stipulated and limited, the terms "installed", "provided with", "mounted / connected", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection. It can be a mechanical connection or an electrical connection. It can be a direct connection or an indirect connection through an intermediate medium. It can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0063] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, step identifiers S101, S102, S103, etc. are only used for convenient expression and do not represent the execution order, and the corresponding execution order can be adjusted.

[0064] See also Figure 1 As shown, this embodiment provides a method for predicting face aging, including:

[0065] S101, receiving an input face image, real age, predicted age and maintenance frequency; the input image includes at least one;

[0066] S102, analyzing the real skin condition based on the face image;

[0067] S103, calculating an aging factor based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition;

[0068] S104, calculating a maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, hydration frequency, and staying up late frequency, and the maintenance factor is inversely correlated with the sun protection frequency and the hydration frequency, and is positively correlated with the staying up late frequency;

[0069] S105, calculating an aging coefficient based on the real age, the predicted age, the aging factor and the maintenance factor; wherein the aging coefficient is positively correlated with the aging factor, the maintenance factor and the difference between the predicted age and the real age.

[0070] The execution subject of the above S101 to S105 can be an electronic device with a facial image acquisition component, especially an electronic device that can obtain clear images under different light sources, such as a server, a mobile phone, a tablet computer or a proprietary electronic device, etc. The input facial image can be collected in real time by the electronic device, or it can be collected and stored by the electronic device. In addition, the facial image can also be collected by a collection component that can obtain clear images under different light sources and sent to the electronic device. It can be set specifically as needed, and this embodiment does not limit it.

[0071] Furthermore, in S105, after the aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor, the relationship diagram between the aging coefficient and age can be displayed on the electronic device that executes S101 to S105, or sent to other display terminals for display so that users can directly view it. Of course, the aging coefficient can also be saved only in the electronic device for subsequent calculations. It can be set specifically as needed, and this embodiment does not limit it.

[0072] Specifically, the aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor, as shown below:

[0073] F=a^(z-x+b)

[0074] Among them, F represents the aging coefficient; x represents the real age; z represents the predicted age; a represents the aging factor; and b represents the maintenance factor.

[0075] See also Figure 2 The figure shows the relationship between the aging coefficient and age of different users. It can be seen that the larger the aging factor and maintenance factor, the faster the aging speed.

[0076] In the above formula, the aging factor a is related to the real age x and the real skin condition coefficient c. The optimal value range of a is 1.02-1.04. The larger a is, the faster the aging. b is mainly related to the sun protection factor V and the hydration coefficient W. The optimal value range of b is (0,20). The value of b is inversely related to V and W, and positively related to the staying up late coefficient T. The larger V and W are, the smaller b is. The smaller the staying up late coefficient, the smaller b is, and the slower the aging speed is.

[0077] Furthermore, the aging factor is calculated based on the real skin condition, which is expressed as follows:

[0078] a=a1+a2*(1+c)

[0079] Wherein, a represents the aging factor; a1 and a2 are constants, a1 can be a constant near 1, and a2 can be a constant near 0.01; c represents the real skin condition coefficient.

[0080] The real skin condition is analyzed based on the face image, as shown below:

[0081]

[0082] Among them, c represents the real skin condition coefficient; x represents the real age; fn represents the score of the skin features extracted based on the face image, increasing from 0 to 1; N represents the number of skin features.

[0083] Some features are not easy to observe under white light, but they are strongly related to the aging of facial skin. By taking facial images under different light sources, facial skin features are extracted and used for aging prediction.

[0084] The various light sources of the present invention include: white light, cross-polarized light, parallel polarized light, ultraviolet light, blue light, etc. By using different light sources, facial images under different light sources are obtained by shooting with a camera.

[0085] Specifically, skin features include white light spots, wrinkles, brown spots, ultraviolet spots, and pigmentation, etc. The white light spots are acquired based on facial images collected under white light; the wrinkles are acquired based on facial images collected under cross-polarized light; the brown spots are acquired based on facial images collected under parallel polarized light; the ultraviolet spots are acquired based on facial images collected under ultraviolet light; and the pigmentation is acquired based on facial images collected under blue light.

[0086] The specific method of performing facial feature detection on facial images with different light sources and extracting the score of skin features is as follows.

[0087] The mask mask1 of white light spots is extracted from the white light face image. The score f1 of white light spots is obtained by calculating the area ratio of the white light spot mask1 in the face skin. Specifically, the severity of spots is divided into four categories: none, mild, moderate, and severe. The area ratio is [0, 0.005] for none, [0.005, 0.015] for mild, [0.015, 0.03] for moderate, and [0.03, 1] for severe. The scores of the four categories are [0.2, 0.4, 0.6, 0.8]. The same calculation method as white light spots is used to obtain the mask and score f2 of brown spots mask2 from the cross-polarized light face imaging; the mask mask3 and score f3 of ultraviolet spots are obtained from the ultraviolet light face imaging; and the mask and score f4 of pigmentation mask4 are obtained from the blue light face image. The wrinkle mask mask5 and score f5 can also be obtained from the face imaging of parallel polarized light. The wrinkles can be divided into four categories according to the area using the same calculation method as the white light spots (the area can comprehensively reflect the number and thickness of wrinkles), [0, 0.002] for none, [0.001, 0.006] for light, [0.006, 0.01] for medium, and [0.01, 1] for heavy. The scores of the four categories are [0.2, 0.4, 0.6, 0.8].

[0088] The classification in the present invention is based on data statistics so that the number of faces in each category can present a normal distribution.

[0089] The score fn of each skin feature ranges from 0 to 1. When the score fn is close to 0, the current feature situation is close to the feature performance of a newborn, and when the feature is close to 1, it indicates that the current feature situation is close to the feature performance of a 100-year-old. The aging speed coefficient is calculated by f1, f2, f3, f4, and f5. The larger the c, the faster the aging speed. When c<0, it indicates that the aging speed is lower than the normal level. When c=0, it indicates that the aging speed is at the normal level. When c>0, it indicates that the aging speed is higher than the normal level.

[0090] Furthermore, the maintenance factor is calculated based on the maintenance frequency, which is expressed as follows:

[0091]

[0092] Among them, b represents the maintenance factor; b1 and b1 are constants, b1 can be a constant around 10~20, b2 can be a constant between 0 and 5; V represents the frequency of sun protection; W represents the frequency of hydration; T represents the frequency of staying up late.

[0093] See also Figure 3 As shown, after the aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor, it also includes:

[0094] S106, predicting the skin characteristics of the predicted age according to the aging coefficient to obtain predicted skin characteristics; the predicted skin characteristics at least include spots and wrinkles;

[0095] S107, displaying the predicted skin features on the white light face image.

[0096] Furthermore, the skin characteristics of the predicted age are predicted according to the aging coefficient to obtain the predicted skin characteristics, which specifically include:

[0097] Extracting masks of skin features related to predicted skin features from face images acquired under different light sources;

[0098] Adding and / or superimposing masks of relevant skin features to obtain a predicted skin feature mask;

[0099] A convolution operation is performed on the predicted feature mask and a convolution kernel related to the aging coefficient, and channel processing is performed based on the predicted feature mask after the convolution operation to obtain the predicted skin features.

[0100] The following is an explanation using the example of predicting skin features as spots.

[0101] As above, a white light spot mask mask1 is obtained by a white light image, a brown spot mask mask2 is obtained by a cross-polarized light image, an ultraviolet spot mask mask3 is obtained by an ultraviolet light image, and a pigmentation mask mask4 is obtained by a blue light image. A spot mask mask is obtained based on mask1, mask2, mask3, and mask4.

[0102] Specifically, the non-overlapping areas of mask1, mask2, mask3, and mask4 are directly added together. In areas where multiple color spot masks overlap, they are multiplied by a factor of 0 to 1 after adding. The purpose of this is to increase the weight in areas where different types of color spots appear repeatedly without making the value of the color spot mask too large.

[0103] Furthermore, a convolution kernel conv_kernel is set to smooth the edges of the color spots of the color spot mask mask obtained above and to deepen the color spot mask. The conv_kernel is related to the aging coefficient F used for prediction and can be obtained by multiplying F with a preset Gaussian convolution kernel.

[0104] The preset Gaussian convolution kernel is a convolution kernel with a sum of 1. Therefore, when the value of F is 1, the convolution kernel only performs a smoothing operation on the mask. When the value of F is greater than 1, the larger the value of F, the more obvious the deepening effect on the mask.

[0105]

[0106] By using conv_kernel to perform convolution operation on mask, the edge of the color spot is smoothed and the color spot mask is deepened and widened to obtain mask_conv.

[0107] mask_conv=conv(mask,conv_kernel,stride,padding)

[0108] Here, conv() represents the convolution operation, tride represents the step size of the convolution, which can be set to 1 in the application, and padding represents whether to fill back to the mask size, which can be set to true in the application.

[0109] Use the color spot mask mask_conv to predict color spot aging for white light imaging. It is necessary to process the three color channels of white light imaging to make the color spot prediction effect more natural.

[0110] img_pre_r=img_r–k_r*mask_conv

[0111] img_pre_g=img_g–k_g*mask_conv

[0112] img_pre_b=img_b–k_b*mask_conv

[0113] Among them, img_pre_r represents the predicted white light r channel image, img_pre_g represents the predicted white light g channel image, and img_pre_b represents the predicted white light b channel image; img_r represents the white light r channel image, img_g represents the white light g channel image, and img_b represents the white light b channel image; k_r represents the prediction coefficient of the white light r channel, k_g represents the prediction coefficient of the white light g channel, and k_b represents the prediction coefficient of the white light b channel.

[0114] Here, k_r, k_g and k_b can change according to the skin color of the face, and the sum of the three is 1, so that a more natural effect can be maintained on different faces.

[0115] Specifically, the white light image can be passed into the face skin segmentation model to obtain the area containing only the face skin, and the mean of each channel in this area is calculated to obtain the mean r_mean of the r channel, the mean g_mean of the g channel, and the mean b_mean of the b channel, and then:

[0116] k_r=r_mean / (r_mean+g_mean+b_mean)

[0117] k_g=g_mean / (r_mean+g_mean+b_mean)

[0118] k_b=b_mean / (r_mean+g_mean+b_mean)

[0119] Finally, the image of color spot prediction is obtained by fusing img_pre_r, img_pre_g and img_pre_b.

[0120] See also Figure 4 As shown, this embodiment also discloses a face detection device, including:

[0121] An input data receiving module 401 is used to receive an input face image, real age, predicted age and maintenance frequency; the input image includes at least one;

[0122] A real skin condition acquisition module 402 is used to analyze the real skin condition based on the face image;

[0123] An aging factor calculation module 403 is used to calculate an aging factor based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition;

[0124] The maintenance factor calculation module 404 is used to calculate the maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, water replenishment frequency and staying up late frequency, and the maintenance factor is inversely related to the sun protection frequency and the water replenishment frequency, and is positively related to the staying up late frequency;

[0125] The aging coefficient calculation module 405 is used to calculate the aging coefficient based on the real age, the predicted age, the aging factor and the maintenance factor; wherein the aging coefficient is positively correlated with the aging factor, the maintenance factor and the difference between the predicted age and the real age.

[0126] For further information, see Figure 5 As shown, the face aging prediction device further includes:

[0127] A skin feature prediction module 406, configured to predict the skin features of the predicted age according to the aging coefficient to obtain predicted skin features; the predicted skin features at least include spots and wrinkles;

[0128] The display module 407 is used to display the predicted skin features on the white light face image.

[0129] The specific implementation of each module of a face detection device can be found in a face aging prediction method, which will not be repeated in this embodiment.

[0130] See also Figure 6 FIG. 1 is a schematic diagram of the hardware structure of an electronic device 60 for predicting face aging provided by this embodiment. Figure 6 As shown, the electronic device 60 includes:

[0131] One or more processors 601 and memory 602, Figure 6 A processor 601 is taken as an example.

[0132] The processor 601 and the memory 602 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0133] The memory 602 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the face aging prediction method in the embodiment of the present application. The processor 601 executes various functional applications of the terminal device or server and completes data processing by running the non-volatile software programs, instructions and modules stored in the memory 602, that is, the face aging prediction method in the above method embodiment is implemented.

[0134] The memory 602 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the face aging prediction device, etc. In addition, the memory 602 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 602 may optionally include a memory remotely arranged relative to the processor 601, and these remote memories may be connected to the face aging prediction device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0135] The one or more modules are stored in the memory 602, and when executed by the one or more processors 601, perform the face aging prediction method in any of the above method embodiments, for example, perform the above described Figure 1 Steps 101 to S107 of the method are implemented Figure 4 The functions of modules 401-407 in.

[0136] The above-mentioned product can execute the method provided in the embodiment of the present application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided in the embodiment of the present application.

[0137] The electronic device of the embodiment of the present application exists in various forms, including but not limited to:

[0138] (1) Mobile communication devices: These devices are characterized by their mobile communication functions and their main purpose is to provide voice and data communications. These terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.

[0139] (2) Ultra-mobile personal computer devices: These devices fall into the category of personal computers, have computing and processing capabilities, and generally also have mobile Internet access features. These terminals include: PDA, MID and UMPC devices, such as iPad.

[0140] (3) Server: A device that provides computing services. The server consists of a microcontroller, hard disk, memory, system bus, etc. The server is similar to the general computer architecture, but because it needs to provide highly reliable services, it has higher requirements in terms of processing power, stability, reliability, security, scalability, manageability, etc.

[0141] (4) Other electronic devices with data interaction functions.

[0142] The embodiment of the present application provides a non-volatile computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more microcontrollers, such as Figure 6 A processor 601 in the embodiment may enable the one or more microcontrollers to execute the face aging prediction method in any of the above method embodiments, for example, to execute the above described Figure 1 Steps 101 to S107 of the method are implemented Figure 4 The functions of modules 401-407 in.

[0143] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] Through the description of the above implementation methods, ordinary technicians in this field can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Ordinary technicians in this field can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0145] The above is only a preferred specific implementation of the present invention; however, the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solution and its improved conception within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for predicting facial aging, characterized in that: include: Receive input face image, real age, predicted age and maintenance frequency; The input image includes at least one image; Analyze the real skin condition based on the face image; An aging factor is calculated based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition; Calculate a maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, hydration frequency, and staying up late frequency, and the maintenance factor is inversely correlated with the sun protection frequency and the hydration frequency, and is positively correlated with the staying up late frequency; The aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor; among them, the aging coefficient is positively correlated with the aging factor, maintenance factor and the difference between the predicted age and the real age.

2. The method for predicting facial aging according to claim 1, characterized in that: The aging coefficient is calculated based on the real age, predicted age, aging factor and maintenance factor, and is expressed as follows: F=a^(z-x+b) Among them, F represents the aging coefficient; x represents the real age; z represents the predicted age; a represents the aging factor; and b represents the maintenance factor.

3. The method for predicting facial aging according to claim 1, characterized in that: The aging factor is calculated based on the real skin condition and is expressed as follows: a=a1+a2*(1+c) Among them, a represents the aging factor; a1 and a2 are constants; and c represents the real skin condition coefficient.

4. The method for predicting facial aging according to claim 1, characterized in that: The real skin condition is analyzed based on the face image, as shown below: Among them, c represents the real skin condition coefficient; x represents the real age; fn represents the score of the skin features extracted based on the face image, increasing from 0 to 1; N represents the number of skin features.

5. The method for predicting facial aging according to claim 4, characterized in that: The skin characteristics include at least one of white spots, wrinkles, brown spots, UV spots, and pigmentation.

6. The method for predicting facial aging according to claim 5, characterized in that: The white light spots are acquired based on facial images collected under white light; the wrinkles are acquired based on facial images collected under cross-polarized light; the brown spots are acquired based on facial images collected under parallel polarized light; the ultraviolet spots are acquired based on facial images collected under ultraviolet light; and the pigmentation is acquired based on facial images collected under blue light.

7. The method for predicting facial aging according to claim 4, characterized in that: The skin feature scores are calculated using the following method: Extract the mask corresponding to the skin features from the face image; Calculate the ratio of the area of ​​the mask of the skin feature to the facial skin area in the facial image; The proportion is used as a score of the skin feature.

8. The method for predicting facial aging according to claim 1, characterized in that: The maintenance factor is calculated based on the maintenance frequency and is expressed as follows: Among them, b represents the maintenance factor; b1 and b1 are constants; V represents the frequency of sun protection; W represents the frequency of hydration; and T represents the frequency of staying up late.

9. The method for predicting facial aging according to claim 1, characterized in that: After calculating the aging coefficient based on real age, predicted age, aging factor and maintenance factor, it also includes: Predicting the skin characteristics of the predicted age according to the aging coefficient to obtain predicted skin characteristics; the predicted skin characteristics at least include spots and wrinkles; The predicted skin features are displayed on the white light face image.

10. The method for predicting facial aging according to claim 9, characterized in that: The skin characteristics of the predicted age are predicted according to the aging coefficient to obtain the predicted skin characteristics, including: Extracting masks of skin features related to predicted skin features from face images acquired under different light sources; Adding and / or superimposing masks of relevant skin features to obtain a predicted skin feature mask; A convolution operation is performed on the predicted feature mask and a convolution kernel related to the aging coefficient, and channel processing is performed based on the predicted feature mask after the convolution operation to obtain the predicted skin features.

11. A facial aging prediction device, characterized in that: include: An input data receiving module, used for receiving an input face image, real age, predicted age and maintenance frequency; The input image includes at least one image; A real skin condition acquisition module is used to analyze the real skin condition based on the face image; An aging factor calculation module, used to calculate the aging factor based on the real skin condition; wherein the aging factor is positively correlated with the real skin condition; A maintenance factor calculation module, used to calculate the maintenance factor based on the maintenance frequency; wherein the maintenance frequency includes at least one of sun protection frequency, water replenishment frequency and staying up late frequency, and the maintenance factor is inversely related to the sun protection frequency and the water replenishment frequency, and is positively related to the staying up late frequency; The aging coefficient calculation module is used to calculate the aging coefficient based on the real age, predicted age, aging factor and maintenance factor; wherein the aging coefficient is positively correlated with the aging factor, maintenance factor and the difference between the predicted age and the real age.

12. The face aging prediction device according to claim 11, characterized in that: Also includes: A skin feature prediction module, used to predict the skin features of the predicted age according to the aging coefficient to obtain predicted skin features; the predicted skin features at least include spots and wrinkles; The display module is used to display the predicted skin features on the white light face image.

13. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores an application program, and the processor is used to run the application program in the memory to execute the steps in the face aging prediction method as described in any one of claims 1 to 10.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps of the face aging prediction method as described in any one of claims 1 to 10.

Images