Muscle nerve tremor time-frequency image processing method and device

Abstract

The application relates to a muscle nerve vibration time-frequency image processing method and device, and belongs to the image processing field, wherein the muscle nerve vibration time-frequency image processing method comprises the following steps: collecting a vibration image of a preset muscle; generating a time-frequency image of the preset muscle according to the vibration image of the preset muscle; calling a preset muscle-expression analysis model; and analyzing the psychological type of an object according to the time-frequency image and the muscle-expression analysis model. The application has the effect of inversely deducing the psychological activity of people by combining the analysis of the related information of the muscle and the micro-expression.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing, in particular to a muscle nerve vibration time-frequency image processing method and device. BACKGROUND

[0002] The face of a person can transmit information, it is a medium, an information transmitter; and micro-expression is a kind of information transmitted by the face of a person; people express their inner feelings to the opposite party by making some expressions, and the face leaks other information between different expressions made by people or in a certain expression.

[0003] Micro-expression often flashes by, and even the person making the expression and the observer usually do not perceive the facial expression changes; micro-expression generally contains at least seven kinds, and each expression expresses different meanings, for example, happy, sad, scared, angry, disgusted, surprised and contemptuous.

[0004] For the related technology in the above, the inventor finds that when people make different micro-expressions, different psychological activities in the heart of people are represented, at this time, the corresponding muscles also change under control, so that the psychological activities of people can be deduced by combining the analysis of the related information of the muscles and micro-expression. SUMMARY

[0005] The present application provides a muscle nerve vibration time-frequency image processing method and device, which has the characteristics of deducing the psychological activities of people by combining the analysis of the related information of the muscles and micro-expression.

[0006] The present application aims to provide a muscle nerve vibration time-frequency image processing method.

[0007] The above application of the present application is achieved by the following technical scheme:

[0008] A muscle nerve vibration time-frequency image processing method, comprising:

[0009] Collecting a vibration image of a preset muscle;

[0010] Generating a time-frequency image of the preset muscle according to the vibration image of the preset muscle;

[0011] Calling a preset muscle-expression analysis model;

[0012] Analyzing the psychological type of the object according to the time-frequency image and the muscle-expression analysis model.

[0013] By adopting the technical scheme, firstly, the vibration image of the preset muscle is collected, then the time-frequency image of the preset muscle can be obtained according to the vibration image, and then the psychological type of the object can be analyzed according to the time-frequency image and the preset muscle-expression analysis model; various emotions can be reflected through the vibration time-frequency and amplitude of the muscle controlled by the brain nerve, and the micro-expression of the face, and then the psychological activity of the object can be inferred through the micro-expression of the face and the time-frequency and amplitude of the muscle controlled by the nerve; in this way, the working efficiency and accuracy of inferring the psychological activity of the object are improved.

[0014] In a preferred example, the step of collecting the vibration image of the preset muscle can further include:

[0015] Collecting sample vibration images of a plurality of collection points of the preset muscle;

[0016] Classifying the sample vibration images according to different positions of the collection points to generate a plurality of types of sample vibration images.

[0017] By adopting the technical scheme, the vibration images corresponding to the collection points at different positions on the muscle are collected, so that the images can more accurately reflect the vibration of the muscle.

[0018] In a preferred example, the positions of the plurality of collection points can include: selecting the middle position, the two end point positions, the quarter position and the three-quarter position of the preset muscle as the collection points.

[0019] By adopting the technical scheme, the corresponding vibration images are collected at the middle position, the two end point positions, the quarter position and the three-quarter position of the muscle, so that the vibration of the muscle can be more accurately reflected.

[0020] In a preferred example, the step of generating the time-frequency image of the preset muscle according to the vibration image of the preset muscle can further include: generating a plurality of types of sample time-frequency images corresponding to a plurality of types of sample vibration images.

[0021] By adopting the technical scheme, the corresponding time-frequency images can be generated according to different types of vibration images, so that the accuracy of the time-frequency image is improved.

[0022] In a preferred example, the step of generating the sample time-frequency image can further include:

[0023] Segmenting the sample vibration image according to a preset time domain to obtain an initial time-frequency image;

[0024] Calculating the frequency difference value of the frequencies corresponding to adjacent time domains in the initial time-frequency image;

[0025] compare the frequency difference value with a preset frequency threshold value;

[0026] if the frequency difference value is greater than the preset frequency threshold value, analyze two frequencies corresponding to the frequency difference value to remove a frequency with an abnormal high frequency;

[0027] if the frequency difference value is less than the preset frequency threshold value, keep the two frequencies corresponding to the frequency difference value;

[0028] obtain a sample time-frequency image by processing the initial time-frequency image through the above process.

[0029] By adopting the above technical solution, the initial time-frequency image is obtained first, then the initial time-frequency image is processed to remove the part with an abnormal frequency in the image, and finally the sample time-frequency image is obtained. In this way, the accuracy of the image is improved.

[0030] In a preferred example, the application can be further configured to generate the sample time-frequency image, and the step includes:

[0031] segment the sample vibration image according to a preset time domain to obtain an initial time-frequency image;

[0032] retrieve a plurality of preset abnormal high frequency images;

[0033] perform similarity matching between the image corresponding to each time domain in the initial time-frequency image and the abnormal high frequency image;

[0034] if the matching is successful, remove the image corresponding to the time domain;

[0035] if the matching fails, keep the image corresponding to the time domain;

[0036] obtain a sample time-frequency image by processing the initial time-frequency image through the above process.

[0037] By adopting the above technical solution, only the images corresponding to different time domain parts in the initial time-frequency image need to be matched with the preset abnormal high frequency image in terms of similarity, so as to determine whether the part of the image is an abnormal frequency. In this way, the sample time-frequency image can be obtained more conveniently and accurately.

[0038] In a preferred example, the application can be further configured to analyze the psychological type of the object according to the time-frequency image and the muscle-expression analysis model, and the step includes:

[0039] obtain psychological type information corresponding to each type according to a plurality of preset sample time-frequency images of the same type of muscle and the muscle-expression analysis model;

[0040] obtain the psychological type of the object by analyzing the psychological type information.

[0041] By adopting the technical scheme, different types of sample time-frequency images of multiple muscles are obtained, and corresponding different types of psychological type information is obtained by cooperating with a muscle-expression analysis model; then, the psychological type of the object can be obtained after analysis; in this way, the psychological type of the object is obtained more efficiently, quickly and accurately.

[0042] The second purpose of the present application is to provide a muscle nerve vibration time-frequency image processing device.

[0043] The second purpose of the present application is achieved by the following technical scheme:

[0044] A muscle nerve vibration time-frequency image processing device comprises:

[0045] A collection module is configured to collect a vibration image of a preset muscle.

[0046] A generation module is configured to generate a time-frequency image of the preset muscle according to the vibration image of the preset muscle.

[0047] A calling module is configured to call a preset muscle-expression analysis model.

[0048] An analysis module is configured to analyze a psychological type of an object according to the time-frequency image and the muscle-expression analysis model.

[0049] The third purpose of the present application is to provide an intelligent terminal.

[0050] The third purpose of the present application is achieved by the following technical scheme:

[0051] An intelligent terminal comprises a memory and a processor, and the memory stores computer program instructions of the muscle nerve vibration time-frequency image processing method which can be loaded and executed by the processor.

[0052] The fourth purpose of the present application is to provide a computer medium which can store corresponding programs.

[0053] The fourth purpose of the present application is achieved by the following technical scheme:

[0054] A computer readable storage medium stores computer programs which can be loaded and executed by a processor to execute any of the muscle nerve vibration time-frequency image processing methods.

[0055] In summary, the present application has at least one of the following beneficial technical effects:

[0056] Through the collection of the vibration image of the muscle, the change of the muscle generated by the brain nerve control under different emotions can be obtained; then the vibration image is processed to obtain a time-frequency image, and then according to the time-frequency image and the muscle-expression analysis model, the psychological type of the object can be obtained, so that in the analysis of the object, the psychological type of the object can be obtained more accurately and quickly. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 Fig. 1 is a flow diagram of a muscle nerve vibration time-frequency image processing method in an embodiment of the present application.

[0058] Figure 2 Fig. 2 is an example diagram of the collection points of different positions of the preset muscle in the muscle nerve vibration time-frequency image processing method in an embodiment of the present application.

[0059] Figure 3 Fig. 3 is a structure diagram of a muscle nerve vibration time-frequency image processing device in an embodiment of the present application.

[0060] The reference signs are explained as follows: 1, collection module; 2, generation module; 3, calling module; 4, analysis module. DETAILED DESCRIPTION

[0061] The present embodiment is only an explanation of the present application, and is not a limitation of the present application. Those skilled in the art can make non-creative modifications to the present embodiment according to the needs after reading the present specification, but as long as the present application is within the scope of the claims, it is protected by the patent law.

[0062] To make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0063] The embodiments of the present application will be described in further detail in combination with the drawings of the present specification.

[0064] The present application provides a muscle nerve vibration time-frequency image processing method, and the main flow of the method is described as follows.

[0065] As shown in the following figure: Figure 1

[0066] Step S101: Collect the vibration image of the preset muscle.

[0067] ​Emotion represents a person's psychological activity, when people have emotion, they will control the muscle through the brain nerve to make the muscle vibrate, that is, the muscle changes, which is reflected to the face as the facial micro-expression; for example, when the emotion is happy, the facial action includes the mouth corner up, the cheek up, the eye lid contraction or the eye tail forming the "fish tail wrinkle" and so on; when the emotion is sad, the facial action includes the eye narrowing, the eyebrow tightening, the mouth corner down or the chin lifting and tightening and so on; when the emotion is afraid, the facial action includes the eye and mouth opening, the eyebrow up or the nostril flaring and so on; when the emotion is angry, the facial action includes the eyebrow down, the forehead wrinkling or the eye lid and the lip tightening and so on; the above facial action all represents the muscle changes, that is, the muscle produces vibration.

[0068] According to the above process, it can be known that when people produce psychological activity, the muscle will vibrate, which further causes the facial micro-expression, and the facial micro-expression can also represent the emotion of people, and the emotion represents the psychological activity of people; therefore, to understand the psychological activity of people, the change of the muscle can be detected; when the muscle changes, it is shown that the muscle vibrates, then the muscle vibration needs to be detected, and the corresponding vibration image is obtained according to the vibration of the muscle in a period of time.

[0069] In the embodiment of the present application, the step of collecting the vibration image of the preset muscle includes: first, collecting the sample point vibration image of a plurality of collection points of the preset muscle; then, classifying the sample point vibration image according to the different positions of the collection points, and finally generating a plurality of types of sample point vibration images.

[0070] In the embodiment of the present application, the above-mentioned plurality of collection points are five collection points; for a plurality of preset muscles, five collection points at different positions are selected, which are the middle position of the preset muscle, the two endpoint positions of the preset muscle, the one fourth position of the preset muscle and the three fourth position of the preset muscle; that is, five collection points at different positions are selected for each muscle, and each position is labeled, and the collection points with the same label on different muscles are the same type of collection points; when collecting the corresponding vibration image of each collection point, the vibration frequency of each collection point is first collected, and then the corresponding vibration image is generated; that is, the vibration image of the preset muscle reflects the vibration frequency of the muscle.

[0071] For example, in the Figure 2 , for muscle A, along the arrow direction, the first endpoint is the first collection point, the one fourth is the second collection point, the middle position is the third collection point, the three fourth is the fourth collection point, and the second endpoint is the fifth collection point; it can be understood that the muscles and the collection points at different positions in the picture are only exemplary.

[0072] The vibration image corresponding to each collection point is referred to as a sample point vibration image, and then the sample point vibration images are classified according to the different collection point labels and positions, for example, the sample point vibration images corresponding to the collection points with label 1 are of the same type, and the sample point vibration images corresponding to the collection points with label 2 are of the same type; in this way, after classification of all the sample point vibration images, a plurality of types of sample point vibration images are obtained.

[0073] Taking five collection points in the embodiment of the present application as an example, each of the preset muscles contains five collection points, and 31 types of sample point vibration images can be obtained after classification and integration of the different collection points; in the classification process, the vibration image generated by the vibration frequency of one collection point is of one type, and five types can be obtained; then the vibration image generated by the vibration frequency of two collection points is of one type, and ten types can be obtained; then the vibration image generated by the vibration frequency of three collection points is of one type, and ten types can be obtained; then the vibration image generated by the vibration frequency of four collection points is of one type, and five types can be obtained; finally, the vibration image generated by the vibration frequency of five collection points is of one type, and one type can be obtained; in summary, if five collection points are taken as an example, 31 types of sample point vibration images can be obtained in total.

[0074] In this way, on the one hand, the vibration images of all the preset muscles are classified and arranged, and the total amount of sample point vibration images can be known, so that the overall workload can be estimated; on the other hand, after classification and arrangement of the vibration images, the next step operation is facilitated, the convenience of image recognition and analysis is improved, and the overall work convenience and work efficiency are improved.

[0075] It can be understood that when collecting the vibration images of the muscles, the vibration images of the muscles can be obtained by detecting at different positions of the collection points through sensors, vibration detectors and the like, and cooperating through mobile terminals and the like; the manner of obtaining the vibration images in the embodiment of the present application is not limited, as long as the vibration images can be obtained.

[0076] Step S102: generating a time-frequency image of the preset muscle according to the vibration image of the preset muscle.

[0077] After obtaining the vibration image of the preset muscle, the vibration image needs to be processed, and then the corresponding time-frequency image is obtained; it can be understood that after obtaining the vibration frequency of the muscle, the corresponding time is obtained, and then the change of the muscle can be shown according to the frequency and time, so that the psychological activity of people can be reflected through the change.

[0078] In the embodiment of the present application, a plurality of types of sample point vibration images can generate a plurality of types of sample point time-frequency images corresponding to the plurality of types.

[0079] In one example, the step of generating the sample time-frequency pattern corresponding to the plurality of types according to the plurality of types of sample vibration images comprises: segmenting the sample vibration images according to a preset time domain to obtain an initial time-frequency image; calculating the frequency difference between the frequencies corresponding to adjacent time domains in the initial time-frequency image to obtain a frequency difference value; comparing the frequency difference value with a preset frequency threshold; if the frequency difference value is greater than the preset frequency threshold, analyzing the two frequencies corresponding to the frequency difference value to remove the frequency with an abnormally high frequency; if the frequency difference value is less than the preset frequency threshold, retaining the two frequencies corresponding to the frequency difference value; and processing the initial time-frequency image through the above process to obtain the sample time-frequency image.

[0080] It can be understood that after obtaining the sample vibration image, the sample vibration image needs to be segmented, and the segmentation is performed according to a preset time domain. The preset time domain is a short time by default, for example, the preset time domain is one minute or two minutes, etc. The specific segmentation process is that, in the sample vibration image, the sample vibration image is segmented according to the preset time domain. Assuming that the sample vibration image is segmented into five segments, the frequency value in each segment is calculated respectively. In the initial time-frequency image, the x-axis represents time, and the y-axis represents the frequency value. The point corresponding to each time point and the corresponding frequency value is a point in the initial time-frequency image. Connecting each point forms the initial time-frequency image.

[0081] After obtaining the initial time-frequency image, the initial time-frequency image needs to be processed. First, the frequencies corresponding to adjacent time domains in the initial time domain image are calculated to obtain a frequency difference value, and then the frequency difference value is compared with a preset frequency threshold. If the frequency difference value is greater than the preset frequency threshold, it indicates that the two frequencies corresponding to the frequency difference value may be abnormal. For example, in the first time domain, the frequency value is 3, and in the second time domain, the frequency value is 15. The frequency difference value corresponding to the two time domains is 12. The preset frequency threshold is 3, and 12>3, which indicates that the two frequencies corresponding to the frequency difference value are abnormal. After analysis, it can be determined that the frequency of the second time domain is an abnormal frequency. The frequency of the second time domain changes too much compared with the frequency of the first time domain, which indicates that the frequency of the second time domain has no reference significance, and thus the image of the second time domain has no reference significance, and the image of the second time domain is removed. If the frequency difference value is less than the preset frequency threshold, it indicates that there is no problem.

[0082] In another example, the step of generating the sample time-frequency image corresponding to the plurality of types according to the plurality of types of sample vibration images comprises: segmenting the sample vibration images according to a preset time domain to obtain an initial time-frequency image; calling a plurality of preset abnormal high-frequency images; performing similarity matching between each time domain corresponding image in the initial time-frequency image and the abnormal high-frequency image; if the matching is successful, removing the time domain corresponding image; if the matching fails, retaining the time domain corresponding image; and obtaining the sample time-frequency image by processing the initial time-frequency image through the above process.

[0083] In the present example, the manner of obtaining the initial time-frequency image is the same as that of the previous example; then a plurality of preset abnormal high-frequency images are called. The abnormal high-frequency image herein can be understood as an image stored in a database in advance. The abnormal high-frequency image represents an abnormal frequency image. When judging the initial time-frequency image, if the image of a certain time domain in the initial time-frequency image has a high similarity with the abnormal high-frequency image, it means that the time domain corresponding image is an abnormal frequency image, and thus the image needs to be removed.

[0084] Through the manners in the above two examples, the generation of the sample time-frequency image can be realized, and the abnormal influence in the image is removed, thereby improving the accuracy of the sample time-frequency image and the work efficiency of the whole work.

[0085] Step S103: calling a preset muscle-expression analysis model.

[0086] It can be understood that the muscle-expression analysis model is preset and stored in a database, and can be directly called for use when needed. Before use, the muscle-expression analysis model needs to be established and trained. In the present application, the muscle-expression analysis model is first established, and then trained. The establishment process of the muscle-expression analysis model comprises: first, crawling historical muscle time-frequency image big data and corresponding psychological activity information according to a crawler, and then training the muscle-expression analysis model according to the historical muscle time-frequency image big data and the psychological activity information, thereby obtaining the trained muscle-expression analysis model. The above process is a common technical means in the related field, and will not be described here.

[0087] Step S104: analyzing the psychological type of the object according to the time-frequency image and the muscle-expression analysis model.

[0088] It can be understood that the muscle-expression analysis model is trained by historical muscle time-frequency image big data and corresponding psychological activity information in the process of training the muscle-expression analysis model. Therefore, after obtaining the sample time-frequency image and the muscle-expression analysis model, the psychological type of the object can also be obtained by analyzing the time-frequency image and the muscle-expression analysis model.

[0089] In the embodiment of the present application, the same type of sample time-frequency images of a plurality of preset muscles are obtained, and corresponding psychological type information of each type is obtained according to a muscle-expression analysis model; then, the psychological types are analyzed to obtain the psychological type of the object.

[0090] In step S101, it can be known that five acquisition points are taken as examples in the embodiment of the present application, and then 31 types of sample vibration images can be obtained; then, 31 types of sample time-frequency images can be obtained according to the 31 types of sample vibration images; then, corresponding 31 types of psychological type information can be obtained according to the muscle-expression analysis model; and then, the 31 types of psychological type information are classified, and the psychological type information with the highest frequency is taken as the psychological type of the object.

[0091] For example, among the 31 types of psychological type information, 18 are happy, 7 are sad, and 6 are contemptuous, so the psychological type of the object is happy.

[0092] Through the processing of steps S101-S104, the psychological activities of people can be inversely deduced by analyzing the related information of muscles and combining micro-expressions.

[0093] The present application also provides a muscle nerve vibration time-frequency image processing device, as shown in Figure 3 The muscle nerve vibration time-frequency image processing device includes an acquisition module 1, a generation module 2, a calling module 3, and an analysis module 4. The acquisition module 1 is used to acquire vibration images of preset muscles; the generation module 2 is used to generate time-frequency images of the preset muscles according to the vibration images of the preset muscles; the calling module 3 is used to call a preset muscle-expression analysis model; and the analysis module 4 is used to analyze the psychological type of an object according to the time-frequency images and the muscle-expression analysis model.

[0094] In order to better execute the program of the above method, the present application also provides an intelligent terminal, which includes a memory and a processor.

[0095] The memory can be used to store instructions, programs, codes, code sets, or instruction sets. The memory can include a storage program area and a storage data area, wherein the storage program area can store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the above muscle nerve vibration time-frequency image processing method, etc.; and the storage data area can store data involved in the above muscle nerve vibration time-frequency image processing method, etc.

[0096] The processor can include one or more processing cores. The processor invokes data stored in the memory by running or executing instructions, programs, code sets or instruction sets stored in the memory, performs various functions and processes data of the present application. The processor can be at least one of an application specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field programmable gate array, a central processing unit, a controller, a microcontroller and a microprocessor. It can be understood that the electronic device for implementing the functions of the processor described above can also be other for different devices, and the embodiments of the present application are not limited specifically.

[0097] The present application also provides a computer readable storage medium, for example, including: a U disk, a mobile hard disk, a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes. The computer readable storage medium stores a computer program capable of being loaded by the processor and executing the muscle nerve vibration time-frequency image processing method described above.

[0098] The above description is only the preferred embodiments of the present application and the explanation of the applied technical principles. Those skilled in the art should understand that the disclosure range involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and also covers other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the technical solutions formed by replacing the above features with the technical features disclosed in the present application (but not limited to) having similar functions.

Claims

1. A method for processing time-frequency images of muscle and nerve vibrations, characterized in that, include: Acquire vibration images of preset muscles; Generate a time-frequency image of the preset muscle based on the vibration image of the preset muscle; Retrieve a preset muscle-expression analysis model; The psychological type of the object is analyzed based on the time-frequency images and muscle-expression analysis model; The step of acquiring vibration images of a preset muscle includes: Collect sample vibration images of several preset muscle acquisition points; The sample point vibration images are classified according to the different locations of the acquisition points to generate multiple types of sample point vibration images; The locations of the aforementioned collection points include: selecting the middle position, two endpoint positions, one-quarter position, and three-quarter position of a preset muscle as collection points; selecting five collection points for each muscle and labeling each position; collection points with the same label on different muscles are collection points of the same type. After integrating and classifying the data from different collection points, 31 types of sample point vibration images can be obtained. During the classification process, if the vibration image generated by the vibration frequency of one acquisition point is considered as one type, five types can be obtained; if the vibration image generated by the vibration frequency of two acquisition points is considered as one type, ten types can be obtained; if the vibration image generated by the vibration frequency of three acquisition points is considered as one type, ten types can be obtained; if the vibration image generated by the vibration frequency of four acquisition points is considered as one type, five types can be obtained; and if the vibration image generated by the vibration frequency of five acquisition points is considered as one type, one type can be obtained. The step of generating a time-frequency image of a preset muscle based on the vibration image of the preset muscle includes generating a time-frequency image of a sample point corresponding to a variety of types based on the vibration images of the sample points of the preset muscle. The steps for generating a sample time-frequency image include: The initial time-frequency image is obtained by segmenting the sample vibration image according to the preset time domain. The frequency difference is obtained by calculating the frequencies corresponding to adjacent time domains in the initial time-frequency image; The frequency difference is compared with a preset frequency threshold. If the frequency difference is greater than a preset frequency threshold, the two frequencies corresponding to the frequency difference are analyzed and abnormally high frequencies are removed. If the frequency difference is less than a preset frequency threshold, then the two frequencies corresponding to the frequency difference are retained. The initial time-frequency image is processed through the above process to obtain the sample time-frequency image; Alternatively, the steps for generating a sample time-frequency image include: The initial time-frequency image is obtained by segmenting the sample vibration image according to the preset time domain. Retrieve multiple preset abnormal high-frequency images; Perform similarity matching between the image corresponding to each time domain in the initial time-frequency image and the abnormal high-frequency image; If a match is found, the image corresponding to that time domain is removed. If a match fails, the image corresponding to that time domain is retained; The initial time-frequency image is processed through the above process to obtain the sample time-frequency image.

2. The muscle and nerve vibration time-frequency image processing method according to claim 1, characterized in that, The steps for analyzing the psychological type of the object based on the time-frequency image and muscle-expression analysis model include: Based on the time-frequency images of sample points of the same type of muscle from multiple preset muscles and the muscle-expression analysis model, we obtain the corresponding psychological type information for each type. The psychological type of the object is obtained by analyzing the psychological type information.

3. A muscle nerve vibration time-frequency image processing device based on the muscle nerve vibration time-frequency image processing method according to claim 1 or 2, characterized in that, include: Acquisition module (1) is used to acquire vibration images of preset muscles; The generation module (2) is used to generate a time-frequency image of a preset muscle based on the vibration image of the preset muscle; The retrieval module (3) is used to retrieve a preset muscle-expression analysis model; The analysis module (4) is used to analyze the psychological type of the object based on the time-frequency image and the muscle-expression analysis model.

4. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores computer program instructions that can be loaded by the processor and executed according to any one of the methods of claims 1-2.

5. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed according to any one of the methods of claims 1-2.

Images