Facial image analysis method and system, electronic equipment and storage medium

Abstract

The invention provides a facial image analysis method and system, electronic equipment and a storage medium, and relates to the field of artificial intelligence, in particular to the fields of big data, image processing and the like, and the specific implementation scheme is as follows: the analysis method comprises the following steps: based on sample facial images of each target sample user under different expressions, obtaining a facial image; acquiring a plurality of groups of sample facial feature data and corresponding sample facial muscle associated data; wherein the sample facial muscle associated data comprises sample facial muscle data used under corresponding expressions and corresponding sample muscle development degree data; constructing a target data conversion relationship according to each group of sample facial feature data and the corresponding sample facial muscle associated data; obtaining a target face image of a target user and corresponding target face feature data; and based on the target facial feature data and the target data conversion relationship, obtaining an analysis result of the target facial image.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence, and in particular to the fields of big data and image processing. Background Art

[0002] Physical health has always been a focus of people's attention. There are many mature solutions for monitoring health status. However, the existing health status monitoring methods are generally one-sided, subjective, and unreliable, and cannot meet the monitoring needs of users. Summary of the invention

[0003] The technical problem to be solved by the present disclosure is to overcome the above-mentioned defects in the prior art and to provide a facial image analysis method, system, electronic device and storage medium.

[0004] The present invention solves the above technical problems through the following technical solutions:

[0005] According to one aspect of the present disclosure, a facial image analysis method is provided, the analysis method comprising:

[0006] Based on sample facial images of each target sample user under different expressions, several groups of sample facial feature data and corresponding sample facial muscle association data are obtained;

[0007] The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data;

[0008] Constructing a target data conversion relationship according to each group of the sample facial feature data and the corresponding sample facial muscle association data;

[0009] Obtaining a target facial image of a target user and corresponding target facial feature data;

[0010] Based on the target facial feature data and the target data conversion relationship, an analysis result of the target facial image is obtained.

[0011] According to another aspect of the present disclosure, a facial image analysis system is provided, the analysis system comprising:

[0012] A sample data acquisition module, used to acquire several groups of sample facial feature data and corresponding sample facial muscle association data based on sample facial images of each target sample user under different expressions;

[0013] The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data;

[0014] A conversion relationship building module, used to build a target data conversion relationship according to each group of the sample facial feature data and the corresponding sample facial muscle association data;

[0015] A target feature data acquisition module is used to acquire a target facial image of a target user and corresponding target facial feature data;

[0016] The image analysis module is used to obtain the analysis result of the target facial image based on the target facial feature data and the target data conversion relationship.

[0017] According to another aspect of the present disclosure, there is provided an electronic device, including:

[0018] at least one processor; and

[0019] a memory communicatively connected to the at least one processor; wherein,

[0020] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor so that the at least one processor can perform the above method.

[0021] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the above method.

[0022] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein the computer program implements the above method when executed by a processor.

[0023] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0025] Figure 1 is a first schematic diagram of a facial image analysis method according to a first embodiment of the present disclosure;

[0026] Figure 2 is a second schematic diagram of the facial image analysis method according to the first embodiment of the present disclosure;

[0027] Figure 3 is a third schematic diagram of the facial image analysis method according to the first embodiment of the present disclosure;

[0028] Figure 44 is a module diagram of a facial image analysis system according to the first embodiment of the present disclosure.

[0029] Figure 5 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0031] Physical health has always been a focus of people's attention. There are many mature solutions for monitoring health status. Traditional methods are mainly implemented in the following ways: (1) based on heart rate data analysis, obtain stress index; (2) based on questionnaire analysis; however, in method (1), the health stress situation based on heart rate analysis is still relatively one-sided. The heart rate situation is more closely related to the user's real-time exercise status. The health and stress index statistics are more indirectly measured and have low credibility and can only be used as a reference. In method (2), the use of questionnaires for health evaluation is relatively traditional. Users can fake their real thoughts through adversarial training, and then fake the evaluation of their mental state. Moreover, the questionnaire questions themselves are highly subjective and are not truly scientific numerical quantifications. The conclusions can only be used as a reference. Therefore, the existing methods cannot achieve accurate and effective analysis of the user's health status.

[0032] In response to the above-mentioned existing problems, the present disclosure proposes a method of using facial muscles directly as indicator data, and analyzing the usage of facial muscles to obtain expression usage, so as to ultimately determine the user's health status. Compared with the above-mentioned analysis results based on heart rate and questionnaires, etc., it is more accurate and reliable.

[0033] Example 1

[0034] like Figure 1 As shown, the facial image analysis method of this embodiment includes:

[0035] S101, based on sample facial images of each target sample user under different expressions, obtaining several groups of sample facial feature data and corresponding sample facial muscle association data;

[0036] The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data;

[0037] Specifically, different facial expressions include crying, laughing, sadness, and expressionless, etc.; different facial muscles are activated in different expression states; through manual and unsupervised training, the expression type corresponding to each sample facial image under different expressions, which facial muscles are activated, and the degree of development of these facial muscles are calibrated. The degree of development of facial muscles can be divided into multiple levels, such as muscle atrophy (01), underdeveloped muscles (02), average development (03), normal development (04), and overdevelopment (05), and these data are stored in a preset database.

[0038] In addition, the data with poor recognition are cleaned to ensure the computational complexity, processing efficiency and image analysis accuracy of subsequent data processing.

[0039] S102, constructing a target data conversion relationship according to each group of sample facial feature data and corresponding sample facial muscle association data;

[0040] Among them, the target data conversion relationship includes preset models, mapping relationships, etc.

[0041] S103, obtaining a target facial image of a target user and corresponding target facial feature data;

[0042] S104: Based on the target facial feature data and the target data conversion relationship, obtain an analysis result of the target facial image.

[0043] The analysis results include analyzing the target users' usage of different expressions, health conditions, etc. in their daily lives in the past.

[0044] In this solution, the muscle distribution of the target user's face and the degree of development of different facial muscles in any facial image can be automatically extracted, and then the analysis content representing the target user's emotions or body condition can be automatically analyzed, including the analysis of the target user's use of different expressions in daily life in the past, and then the target user's stress index, mood index, health index and other health status indicators can be analyzed, thereby providing users with reliable suggestions for improving healthy living and enhancing the user experience.

[0045] In one feasible solution, Figure 1 Further improvements to the above scheme, specifically:

[0046] like Figure 2 As shown, step S104 includes:

[0047] S201, acquiring target facial muscle association data of the target user's face based on the target facial feature data and the target data conversion relationship;

[0048] The target facial muscle associated data includes the target facial muscle data in the target facial image and the corresponding target muscle development degree data;

[0049] S202: Analyze and obtain usage degree data of different expressions of the target user in the target facial image based on the target facial muscle association data.

[0050] In this solution, the target facial muscle data and the corresponding target muscle development data that match the target facial feature data are automatically analyzed through the target data conversion relationship, that is, each facial muscle of the target user's face and its corresponding development level in the target facial image are automatically identified to determine the target user's usage of each expression such as crying and laughing, and then the target user's usage of different expressions in daily life in the past can be determined, such as whether he is a person who laughs often or cries often, etc., so as to realize timely and reliable acquisition of the user's emotional dynamic changes.

[0051] In an implementable solution, after step S104, the method further includes:

[0052] Based on the usage data of different expressions, the health status of the target users can be analyzed.

[0053] In this solution, based on determining the target user's usage of different expressions, the health status of the target user is further analyzed, including stress index, mood index, health index and other health status indicators, so as to provide users with targeted and reliable health life improvement suggestions and enhance the user experience.

[0054] In one feasible solution, step S102 includes:

[0055] Taking each group of sample facial feature data as input and the corresponding sample facial muscle association data as output, the preset model is trained to construct a facial muscle prediction module;

[0056] There is no specific requirement for the model type of the preset model, as long as it can achieve the effect of facial muscle prediction in actual scenes.

[0057] Step S201 includes:

[0058] The target facial feature data is input into the facial muscle prediction module to output the target facial muscle association data of the target user's face.

[0059] In this solution, a facial muscle prediction module is trained based on several historical sample data to predict the facial muscles of the user and their muscle development level in any facial image. That is, the deep learning model is relied upon to analyze the target facial image of the target user, thus ensuring the accuracy and reliability of image processing.

[0060] In one feasible solution, step S102 includes:

[0061] Constructing a corresponding mapping relationship between each set of sample facial feature data and the corresponding sample facial muscle association data and storing the mapping relationship in a preset database;

[0062] Step S201 includes:

[0063] Based on the target facial feature data and the mapping relationship, a search is performed in a preset database to obtain target facial muscle association data that matches the target facial feature data.

[0064] In this solution, through a large amount of sample data, relying on the one-to-one correspondence between each group of sample facial feature data and the corresponding sample facial muscle association data, a mapping table corresponding to the facial muscle association data that matches different facial feature data is constructed; for the target facial feature data of any target user, the corresponding facial muscle association data can be obtained by directly searching the mapping table, thereby ensuring the feasibility and efficiency of the target facial image analysis of the target user, and also ensuring the accuracy and reliability of image processing.

[0065] In one feasible solution, step S202 includes:

[0066] Based on the pre-constructed expression usage degree calculation formula and the target facial muscle association data, the target user's usage degree data of each expression is calculated; wherein each expression corresponds to an expression usage degree calculation formula.

[0067] In this scheme, for any target user's target facial image, based on the target user's facial muscles in the target facial image and their corresponding muscle development levels, they are input into the usage degree calculation formula for each expression, and the probability of the target user using each expression in daily life can be automatically calculated, so as to achieve accurate analysis of the target user's emotional state and facial state.

[0068] In one feasible solution, Figure 3 As shown, the steps of analyzing and obtaining the health status of the target user based on the usage degree data of different expressions include:

[0069] S301, obtaining standard proportion data of each expression used by a standard healthy user at the same time;

[0070] Among them, standard facial images of several standard healthy users in any state are obtained, and each muscle in the face and its corresponding muscle development level are determined according to the standard facial images. Then, the expression usage degree calculation formula is used to pre-calculate the initial proportion data of each standard healthy user using each expression at the same time, and then the statistical methods such as mean value, variance, and chi-square are used to calculate the final standard proportion data of each expression used by the standard healthy user at the same time; for example, the usage ratio of the expression of smile to the expression of cry of the standard healthy user is 4:1.

[0071] S302, determining actual proportion data of each expression used by the target user in the target facial image according to the usage degree data of different expressions;

[0072] S303: Determine the health status of the target user based on the difference between the actual ratio data and the standard ratio data.

[0073] For example, the calculation process of the usage degree of the two expressions of smile and cry on the target user's face is as follows:

[0074] Expression smile gain n1 = 3*frontalis muscle gain + 1*occipitalis muscle gain + 2*levator labii superioris muscle gain + 2*orbicularis oris muscle gain + 3*zygomatic muscle gain...

[0075] Expression crying gain n2 = 1*frontalis muscle gain + 2*occipital muscle gain + 1*orbicularis oris muscle gain +...

[0076] Frontalis muscle*(expression smile gain+expression cry gain)=standard coefficient*occipitalis muscle*(expression smile gain+expression cry gain)

[0077] n1*frontalis gain + n2*frontalis gain = (frontalis gain & occipitalis standard coefficient) * (n1*occipitalis gain + n2*occipitalis gain);

[0078] Among them, the frontal muscle of the standard healthy user = 0.7*occipital muscle, the frontal muscle of the actual target user = 1.2*occipital muscle, frontal muscle gain, occipital muscle gain, ..., muscle gain is the influence factor of the development degree of the corresponding muscle on the expression, and the development degree of the muscle is positively correlated with the influence factor, the expression smile gain is the degree of use of the expression smile, and the expression cry gain is the degree of use of the expression cry;

[0079] Therefore, 3*n1+1*n2=1.2*(1*n1+2*n2), that is, 3n1=2n2;

[0080] It can be determined that the target user's smiling and crying expressions usage ratio n1:n2 is 2:3, while the standard healthy user's smiling and crying expressions usage ratio is 4:1. It can be concluded that this user has more crying expressions than the standard healthy user. The current mental condition is not ideal and targeted adjustments are needed to improve the mental state.

[0081] In this solution, based on the analysis of the target user's facial muscle usage, the target user's use of different expressions is further accurately analyzed to determine the emotional changes, so as to obtain analysis reports on the target user's health, stress, etc., remind and guide the user to make improvements, and enhance the user experience.

[0082] In an implementable solution, before step S101, the method further includes:

[0083] A number of initial sample users are grouped according to preset division conditions to obtain target sample users in different groups;

[0084] Among them, the preset division conditions include the same age group and / or the same gender.

[0085] In this solution, considering different genders, different age groups, etc., there will be a certain degree of difference in the facial muscle status. Therefore, in the sampling stage, based on influencing parameters such as gender and age, several initial sample users are grouped and processed, and the above-mentioned execution logic is used to perform image analysis processing on each group, so as to further improve the analysis accuracy of the health status of the target user in any target user image.

[0086] Example 2

[0087] like Figure 4 As shown, the facial image analysis system of this embodiment includes:

[0088] The sample data acquisition module 401 is used to acquire several groups of sample facial feature data and corresponding sample facial muscle association data based on sample facial images of each target sample user under different expressions;

[0089] The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data;

[0090] Specifically, different facial expressions include crying, laughing, sadness, and expressionless, etc.; different facial muscles are activated in different expression states; through manual and unsupervised training, the expression type corresponding to each sample facial image under different expressions, which facial muscles are activated, and the degree of development of these facial muscles are calibrated. The degree of development of facial muscles can be divided into multiple levels, such as muscle atrophy (01), underdeveloped muscles (02), average development (03), normal development (04), and overdevelopment (05), and these data are stored in a preset database.

[0091] In addition, the data with poor recognition are cleaned to ensure the computational complexity, processing efficiency and image analysis accuracy of subsequent data processing.

[0092] The conversion relationship building module 402 is used to build a target data conversion relationship according to each set of sample facial feature data and the corresponding sample facial muscle association data;

[0093] Among them, the target data conversion relationship includes preset models, mapping relationships, etc.

[0094] The target feature data acquisition module 403 is used to acquire the target facial image of the target user and the corresponding target facial feature data;

[0095] The image analysis module 404 is used to obtain the analysis result of the target facial image based on the target facial feature data and the target data conversion relationship.

[0096] The analysis results include analyzing the target users' usage of different expressions, health conditions, etc. in their daily lives in the past.

[0097] In this solution, the muscle distribution of the target user's face and the degree of development of different facial muscles in any facial image can be automatically extracted, and then the analysis content representing the target user's emotions or body condition can be automatically analyzed, including the analysis of the target user's use of different expressions in daily life in the past, and then the target user's stress index, mood index, health index and other health status indicators can be analyzed, thereby providing users with reliable suggestions for improving healthy living and enhancing the user experience.

[0098] In one feasible solution, Figure 1 Further improvements to the above scheme, specifically:

[0099] The image analysis module 404 includes:

[0100] A target muscle data acquisition unit, used for acquiring target facial muscle associated data of a target user's face based on the target facial feature data and the target data conversion relationship;

[0101] The target facial muscle associated data includes the target facial muscle data in the target facial image and the corresponding target muscle development degree data;

[0102] The image analysis unit is used to analyze and obtain the usage degree data of different expressions of the target user in the target facial image according to the target facial muscle association data.

[0103] In this solution, the target facial muscle data and the corresponding target muscle development data that match the target facial feature data are automatically analyzed through the target data conversion relationship, that is, each facial muscle of the target user's face and its corresponding development level in the target facial image are automatically identified to determine the target user's usage of each expression such as crying and laughing, and then the target user's usage of different expressions in daily life in the past can be determined, such as whether he is a person who laughs often or cries often, etc., so as to realize timely and reliable acquisition of the user's emotional dynamic changes.

[0104] In one feasible solution, the analysis system further includes:

[0105] The health analysis module is used to analyze the health status of the target user based on the usage data of different expressions.

[0106] In this solution, based on determining the target user's usage of different expressions, the health status of the target user is further analyzed, including stress index, mood index, health index and other health status indicators, so as to provide users with targeted and reliable health life improvement suggestions and enhance the user experience.

[0107] In an implementable solution, the conversion relationship building module 402 is further used to take each set of sample facial feature data as input and the corresponding sample facial muscle association data as output, train a preset model, and construct a facial muscle prediction module;

[0108] There is no specific requirement for the model type of the preset model, as long as it can achieve the effect of facial muscle prediction in actual scenes.

[0109] The target feature data acquisition module 403 is used to input the target facial feature data into the facial muscle prediction module to output the target facial muscle association data of the target user's face.

[0110] In this solution, a facial muscle prediction module is trained based on several historical sample data to predict the facial muscles of the user and their muscle development level in any facial image. That is, the deep learning model is relied upon to analyze the target facial image of the target user, thus ensuring the accuracy and reliability of image processing.

[0111] In an implementable solution, the conversion relationship construction module 402 is further used to construct a corresponding mapping relationship between each set of sample facial feature data and the corresponding sample facial muscle association data and store it in a preset database;

[0112] The target feature data acquisition module 403 is further used to search in a preset database based on the target facial feature data and the mapping relationship to obtain target facial muscle association data matching the target facial feature data.

[0113] In this solution, through a large amount of sample data, relying on the one-to-one correspondence between each group of sample facial feature data and the corresponding sample facial muscle association data, a mapping table corresponding to the facial muscle association data that matches different facial feature data is constructed; for the target facial feature data of any target user, the corresponding facial muscle association data can be obtained by directly searching the mapping table, thereby ensuring the feasibility and efficiency of the target facial image analysis of the target user, and also ensuring the accuracy and reliability of image processing.

[0114] In an implementable scheme, the image analysis unit is also used to calculate the usage degree data of each expression of the target user based on a pre-constructed expression usage degree calculation formula and target facial muscle association data; wherein each expression corresponds to an expression usage degree calculation formula.

[0115] In this scheme, for any target user's target facial image, based on the target user's facial muscles in the target facial image and their corresponding muscle development levels, they are input into the usage degree calculation formula for each expression, and the probability of the target user using each expression in daily life can be automatically calculated, so as to achieve accurate analysis of the target user's emotional state and facial state.

[0116] In one feasible solution, the health analysis module includes:

[0117] A standard ratio data acquisition unit, used to acquire standard ratio data of each expression used by a standard healthy user at the same time;

[0118] Among them, standard facial images of several standard healthy users in any state are obtained, and each muscle in the face and its corresponding muscle development level are determined according to the standard facial images. Then, the expression usage degree calculation formula is used to pre-calculate the initial proportion data of each standard healthy user using each expression at the same time, and then the statistical methods such as mean value, variance, and chi-square are used to calculate the final standard proportion data of each expression used by the standard healthy user at the same time; for example, the usage ratio of the expression of smile to the expression of cry of the standard healthy user is 4:1.

[0119] an actual proportion data acquisition unit, used to determine actual proportion data of each expression used by the target user in the target facial image according to the use degree data of different expressions;

[0120] The health analysis unit is used to determine the health status of the target user based on the difference between the actual ratio data and the standard ratio data.

[0121] For example, the calculation process of the usage degree of the two expressions of smile and cry on the target user's face is as follows:

[0122] Expression smile gain n1 = 3*frontalis muscle gain + 1*occipitalis muscle gain + 2*levator labii superioris muscle gain + 2*orbicularis oris muscle gain + 3*zygomatic muscle gain...

[0123] Expression crying gain n2 = 1*frontalis muscle gain + 2*occipital muscle gain + 1*orbicularis oris muscle gain +...

[0124] Frontalis muscle*(expression smile gain+expression cry gain)=standard coefficient*occipitalis muscle*(expression smile gain+expression cry gain)

[0125] n1*frontalis gain + n2*frontalis gain = (frontalis gain & occipitalis standard coefficient) * (n1*occipitalis gain + n2*occipitalis gain);

[0126] Among them, the frontal muscle of the standard healthy user = 0.7*occipital muscle, the frontal muscle of the actual target user = 1.2*occipital muscle, frontal muscle gain, occipital muscle gain, ..., muscle gain is the influence factor of the development degree of the corresponding muscle on the expression, and the development degree of the muscle is positively correlated with the influence factor, the expression smile gain is the degree of use of the expression smile, and the expression cry gain is the degree of use of the expression cry;

[0127] Therefore, 3*n1+1*n2=1.2*(1*n1+2*n2), that is, 3n1=2n2;

[0128] It can be determined that the target user's smiling and crying expressions usage ratio n1:n2 is 2:3, while the standard healthy user's smiling and crying expressions usage ratio is 4:1. It can be concluded that this user has more crying expressions than the standard healthy user. The current mental condition is not ideal and targeted adjustments are needed to improve the mental state.

[0129] In this solution, based on the analysis of the target user's facial muscle usage, the target user's use of different expressions is further accurately analyzed to determine the emotional changes, so as to obtain analysis reports on the target user's health, stress, etc., remind and guide the user to make improvements, and enhance the user experience.

[0130] In one feasible solution, the analysis system further includes:

[0131] A data grouping processing module is used to group a number of initial sample users according to preset division conditions to obtain target sample users in different groups;

[0132] Among them, the preset division conditions include the same age group and / or the same gender.

[0133] In this solution, considering different genders, different age groups, etc., there will be a certain degree of difference in the facial muscle status. Therefore, in the sampling stage, based on influencing parameters such as gender and age, several initial sample users are grouped and processed, and the above-mentioned execution logic is used to perform image analysis processing on each group, so as to further improve the analysis accuracy of the health status of the target user in any target user image.

[0134] Example 3

[0135] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0136] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0137] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0138] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0139] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the above method. For example, in some embodiments, the above method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the above method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the above method in any other appropriate manner (e.g., by means of firmware).

[0140] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0141] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0142] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0143] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0144] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0145] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0146] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0147] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A facial image analysis method, the analysis method comprising: Based on sample facial images of each target sample user under different expressions, several groups of sample facial feature data and corresponding sample facial muscle association data are obtained; The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data; Constructing a target data conversion relationship according to each group of the sample facial feature data and the corresponding sample facial muscle association data; Obtaining a target facial image of a target user and corresponding target facial feature data; Based on the target facial feature data and the target data conversion relationship, an analysis result of the target facial image is obtained.

2. The facial image analysis method according to claim 1, wherein the step of obtaining the analysis result of the target facial image based on the target facial feature data and the target data conversion relationship comprises: Based on the target facial feature data and the target data conversion relationship, acquiring target facial muscle association data of the target user's face; Wherein, the target facial muscle associated data includes the target facial muscle data in the target facial image and the corresponding target muscle development degree data; According to the target facial muscle association data, the usage degree data of different expressions of the target user in the target facial image is analyzed and obtained.

3. The facial image analysis method according to claim 2, further comprising: after the step of analyzing and obtaining the target user's usage degree data of different expressions in the target facial image: The health status of the target user is obtained by analyzing the usage degree data of different expressions.

4. The facial image analysis method according to claim 2, wherein the step of constructing a target data conversion relationship based on each set of the sample facial feature data and the corresponding sample facial muscle association data comprises: Taking each group of sample facial feature data as input and the corresponding sample facial muscle association data as output, training a preset model to construct a facial muscle prediction module; The step of acquiring target facial muscle association data of the target user's face based on the target facial feature data and the target data conversion relationship comprises: The target facial feature data is input into the facial muscle prediction module to output the target facial muscle associated data of the target user's face.

5. The facial image analysis method according to claim 2, wherein the step of constructing a target data conversion relationship based on each set of the sample facial feature data and the corresponding sample facial muscle association data comprises: Constructing a corresponding mapping relationship between each group of the sample facial feature data and the corresponding sample facial muscle association data and storing the mapping relationship in a preset database; The step of acquiring target facial muscle association data of the target user's face based on the target facial feature data and the target data conversion relationship comprises: Based on the target facial feature data and the mapping relationship, the preset database is searched to obtain the target facial muscle association data matching the target facial feature data.

6. The facial image analysis method according to claim 3, wherein the step of analyzing and obtaining the target user's usage degree data of different expressions based on the target facial muscle association data comprises: Based on the pre-constructed expression usage degree calculation formula and the target facial muscle association data, the usage degree data of each expression by the target user is calculated; wherein each expression corresponds to one expression usage degree calculation formula.

7. The facial image analysis method according to claim 6, wherein the step of analyzing the health status of the target user based on the usage data of different expressions comprises: Obtain the standard proportion data of each expression used by a standard healthy user at the same time; Determining actual proportion data of each expression used by the target user in the target facial image according to the usage degree data of different expressions; The health status of the target user is determined based on the difference between the actual ratio data and the standard ratio data.

8. The facial image analysis method according to any one of claims 1 to 7, before the step of acquiring a plurality of groups of sample facial feature data and corresponding sample facial muscle association data based on sample facial images of each target sample user under different expressions, further comprising: Grouping a number of initial sample users according to a preset division condition to obtain the target sample users in different groups; The preset division conditions include the same age group and / or the same gender.

9. A facial image analysis system, the analysis system comprising: A sample data acquisition module, used to acquire several groups of sample facial feature data and corresponding sample facial muscle association data based on sample facial images of each target sample user under different expressions; The sample facial muscle associated data includes the sample facial muscle data used under the corresponding expression and the corresponding sample muscle development degree data; A conversion relationship building module, used to build a target data conversion relationship according to each group of the sample facial feature data and the corresponding sample facial muscle association data; A target feature data acquisition module is used to acquire a target facial image of a target user and corresponding target facial feature data; The image analysis module is used to obtain the analysis result of the target facial image based on the target facial feature data and the target data conversion relationship.

10. The facial image analysis system according to claim 9, wherein the image analysis module comprises: A target muscle data acquisition unit, used for acquiring target facial muscle association data of the target user's face based on the target facial feature data and the target data conversion relationship; Wherein, the target facial muscle associated data includes the target facial muscle data in the target facial image and the corresponding target muscle development degree data; The image analysis unit is used to analyze and obtain the usage degree data of different expressions of the target user in the target facial image according to the target facial muscle association data.

11. The facial image analysis system according to claim 10, further comprising: The health analysis module is used to analyze the health status of the target user based on the usage data of different expressions.

12. The facial image analysis system according to claim 10, wherein the conversion relationship building module is further used to take each group of the sample facial feature data as input and the corresponding sample facial muscle association data as output, train a preset model, and construct a facial muscle prediction module; The target feature data acquisition module is used to input the target facial feature data into the facial muscle prediction module to output the target facial muscle association data of the target user's face.

13. The facial image analysis system according to claim 10, wherein the conversion relationship construction module is further used to construct a corresponding mapping relationship between each group of the sample facial feature data and the corresponding sample facial muscle association data and store the mapping relationship in a preset database; The target feature data acquisition module is also used to search in the preset database based on the target facial feature data and the mapping relationship to obtain the target facial muscle association data matching the target facial feature data.

14. The facial image analysis system according to claim 11, wherein the image analysis unit is further configured to calculate the usage degree data of each expression of the target user based on a pre-constructed expression usage degree calculation formula and the target facial muscle association data; wherein: Each expression corresponds to a calculation formula for the degree of expression usage.

15. The facial image analysis system according to claim 14, wherein the health analysis module comprises: A standard ratio data acquisition unit, used to acquire standard ratio data of each expression used by a standard healthy user at the same time; an actual proportion data acquisition unit, used to determine actual proportion data of each expression used by the target user in the target facial image according to the usage degree data of different expressions; A health analysis unit is used to determine the health status of the target user based on the difference between the actual proportion data and the standard proportion data.

16. The facial image analysis system according to any one of claims 9 to 15, further comprising: A data grouping processing module, used for grouping a number of initial sample users according to a preset division condition to obtain target sample users in different groups; The preset division conditions include the same age group and / or the same gender.

17. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 8.

19. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 8.

Images