Intelligent gymnasium identification management system based on AI (Artificial Intelligence) technology

Abstract

The invention belongs to the technical field of gymnasium recognition management, and particularly discloses a gymnasium intelligent recognition management system based on an AI technology. Comprising a user identity recognition module, a user identity confirmation module, a user behavior recognition module, a user behavior analysis module, a user health recognition module, a user health analysis module, a user coach confirmation module, a database and a gymnasium management terminal. According to the invention, the identity, behavior and health of the user are identified and analyzed, so that insufficient attention to user identification management at present is avoided, personalized service of the user is realized, the pertinence of user guidance is improved, the timeliness of user behavior processing is guaranteed, the use experience of the user is improved, the retention rate of the user is improved, and the user experience is improved. And meanwhile, by analyzing the exercise behavior standard degree and the exercise health standard degree of the target user in each exercise item, the exercise health of the user is guaranteed, so that the occurrence probability of exercise injury of the user is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of fitness gym identification management, and relates to an AI-based fitness gym intelligent identification management system. Background Art

[0002] A gym is a place that provides people with various fitness facilities, venues and related services. It aims to help people enhance their physical fitness, shape a good body, improve their health level and achieve specific fitness goals through various sports and exercises. In order to ensure that users receive professional fitness guidance and high-quality course experience, gyms need to be identified and managed.

[0003] For example, the Chinese invention patent with publication number CN108090688A discloses an intelligent fitness management system, which relates to the field of fitness management technology, and specifically to an intelligent fitness management system. It includes a background system, a management terminal and a venue operation management system, and the background system is data-connected with the management terminal and the venue operation management system. The venue operation management system is used to manage the daily operation of the fitness venue, which includes a user identity recognition system, an access control management system, an environmental management system, and a user data collection system. The user identity recognition system is used to identify the user's identity, and the access control management system is used to manage the entry and exit of people in the gym or fitness cabin. The user data collection system is used to collect user health data, behavior data, and generate user portraits. The background system can generate venue operation recommendation plans based on user portraits. The intelligent fitness management system provided by the present invention can solve the problem that the existing fitness management system cannot quickly and fully mine business data and user data for overall analysis, resulting in inaccurate operation strategies.

[0004] The above existing technologies have the following deficiencies: 1. The current identification and management of fitness gyms mainly focuses on the identification and management of the fitness gyms themselves, that is, insufficient attention is paid to user exercise behaviors and there is a lack of personalized services for users, which leads to insufficient targeted guidance for users. At the same time, it is impossible to deal with user behaviors in a timely manner, which reduces the user's experience and leads to low user retention rate.

[0005] 2. The current analysis of user data mainly focuses on the operational dimension, that is, insufficient attention is paid to the analysis of user behavior and health dimensions, which in turn fails to guarantee the user's sports health, thereby increasing the probability of sports injuries among users. Summary of the invention

[0006] In view of this, in order to solve the problems raised in the above background technology, an intelligent identification management system for a fitness center based on AI technology is proposed.

[0007] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an AI-based intelligent identification and management system for a fitness center, including: a user identity recognition module for identifying the facial image of a target user.

[0008] The user identity confirmation module is used to confirm the identity information of the target user based on the face image of the target user.

[0009] The user behavior recognition module is used to track and shoot the target user's movement process according to the target user's identity information, and then obtain the target user's movement video in various sports events.

[0010] The user behavior analysis module is used to analyze the standardization of the target user's sports behavior in each sports event based on the target user's sports video in each sports event.

[0011] The user health identification module is used to identify the target user's health information in each sports event, including exercise duration and interval duration.

[0012] The user health analysis module is used to analyze the target user's sports health standardization in each sports event based on the target user's health information in each sports event.

[0013] The user coach confirmation module is used to confirm the target user's required coach based on the target user's identity information, the target user's sports behavior norms in various sports events, and the sports health norms.

[0014] The database is used to store the user's permitted exercise time for each sport, store the user information of the fitness center, and store the standard information of each sport in the fitness center.

[0015] The gym management terminal is used to train the target users according to their needs and send instructions to the gym management platform.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention avoids the current problem of insufficient attention to user exercise behavior by identifying and analyzing the user's identity, behavior and health, and realizes personalized service for users, thereby improving the pertinence of user guidance, while ensuring the timeliness of user behavior processing, improving the user's usage experience, and thus improving the user's retention rate.

[0017] (2) The present invention breaks the current lack of attention to the user behavior and health analysis dimensions by analyzing the target user's sports behavior norms and sports health norms in various sports events, thereby ensuring the user's sports health and reducing the probability of sports injuries in the user.

[0018] (3) The present invention confirms the target user's required coach based on the target user's identity information, the target user's sports behavior norms in various sports events, and the sports health norms. This avoids the shortcomings of the current method of confirming the required coach based only on the user's identity level, ensures the user's training safety, and further improves the user's training effect. At the same time, it optimizes the allocation of coach resources, thereby improving the coach's work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 It is a schematic diagram of the connection of each module of the system of the present invention.

[0021] Figure 2 It is a connection diagram of the steps of analyzing the standardization of sports behavior of the present invention.

[0022] Figure 3 It is a connection diagram of the steps of analyzing the sports health standardization of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] See also Figure 1 As shown, the present invention provides a fitness center intelligent identification management system based on AI technology, the system includes: a user identity identification module, a user identity confirmation module, a user behavior identification module, a user behavior analysis module, a user health identification module, a user health analysis module, a user coach confirmation module, a database and a fitness center management terminal.

[0025] In the above, the user identity confirmation module is respectively connected to the user identity recognition module and the user coach confirmation module, the user behavior analysis module is respectively connected to the user behavior recognition module, the user coach confirmation module and the database, the user health analysis module is respectively connected to the user health recognition module, the user coach confirmation module and the database, and the user coach confirmation module is also connected to the fitness center management terminal.

[0026] The user identification module is used to identify the face image of the target user.

[0027] It should be added that the facial image of the target user is identified by using a camera installed in the gym to identify the user's video image, and then capturing the facial image of the target user from the video image.

[0028] The user identity confirmation module is used to confirm the identity information of the target user based on the face image of the target user.

[0029] Exemplarily, confirming the identity information of the target user includes: extracting stored facial images of each user in the gym from a gym management platform.

[0030] The target user’s facial recognition image is matched and compared with the stored facial images of each user in the gym through AI technology.

[0031] When the face recognition image of the target user is consistent with the face image stored in a user in the gym, the identity information of the user is extracted from the user information of the gym as the identity information of the target user.

[0032] When the face recognition image of the target user is inconsistent with the stored face images of each user in the gym, the new user is used as the identity information of the target user.

[0033] The user behavior recognition module is used to track and shoot the target user's movement process according to the target user's identity information, and then obtain the target user's movement video in each sports event.

[0034] It should be added that the sports videos of the target users in various sports are obtained by shooting the cameras of the various sports placed in the fitness center.

[0035] The user behavior analysis module is used to analyze the standardization of the target user's sports behavior in each sports event based on the sports video of the target user in each sports event.

[0036] See also Figure 2 As shown, for example, the analysis of the target user's sports behavior norms in each sports event includes: W1, extracting the joint angles and joint angle change rates of each key action of the target user in each sports event from the sports video of the target user in each sports event, and then calculating the joint norms δ of the target user in each sports event i , i is the sports event number, i=1,2,......,m.

[0037] It should be added that the method for obtaining the joint angles of each key action of the target user in each sports event is as follows: the key action images of the target user in each sports event are extracted from the sports videos of the target user in each sports event, and a suitable coordinate system is established in each key action image, usually with a fixed part of the human body (such as the center of the pelvis) as the origin, and then the coordinates of the joint points in each key image are marked, and the angles between each joint are calculated through vector operations based on the marked joint point coordinates. For example, the calculation of the knee joint angle can be achieved by calculating the angle between the vector from the hip joint to the knee joint and the vector from the knee joint to the ankle joint.

[0038] It should be added that the method of obtaining the rate of change of the joint angle of each key action is as follows: the interval duration between each key action of the target user in each sports event and its next adjacent key action is extracted from the sports video of the target user in each sports event, and the joint angle of each key action and the joint angle of the next adjacent key action are subtracted to obtain the joint angle difference of each key action, and then the ratio of its absolute value to the interval duration is used as the rate of change of the joint angle of each key action.

[0039] Furthermore, the statistical target user's joint standardization in each sports event includes: W1-1, extracting standard information of each sports event in the gym from the database, and then extracting the standard joint angle range of each key action in each sports event.

[0040] W1-2. If the joint angle of a key action of a user in a certain sport is within the standard joint angle range, the key action of the user in the sport is recorded as a qualified key action, and the number of qualified key actions of the sport is counted.

[0041] W1-3, count the number of key actions of the sport, and then use the ratio of the number of qualified key actions of the sport to the number of key actions as the qualified ratio of joint angles of the sport, so as to obtain the qualified ratio of joint angles of the target user in each sport, denoted as τ i .

[0042] In a specific embodiment, the standard joint angle of the hip joint in the squat action in the squat sport event is between 95-115 degrees. When the hip joint angle of the squat action of the target user in the squat sport event is 100 degrees, the squat action is recorded as a qualified squat action, and then the number of qualified squat actions is counted. If the target user involves a total of 50 squat actions in the squat sport event, the number of qualified squat actions is 42.

[0043] The target user's joint angle passing ratio in each sport is statistically = 42÷50 = 0.84.

[0044] W1-4, according to τi The statistical method is similar to the statistical method to obtain the qualified ratio of the target user's joint angle change rate in each sports event, denoted as τ i ′.

[0045] W1-5, τ i and τ i ′ is used to calculate the mean, and the mean calculation result is used as the joint standardization degree of the target user in each sports item, and is recorded as δ i .

[0046] W2, extract the coordinates of the target user's body position center of gravity at each video point in each sport from the target user's sports video in each sport, and then count the target user's body position standardization degree δ in each sport i ′.

[0047] It should be added that the method for obtaining the coordinates of the center of gravity of the target user's body position at each video point in each sports event is as follows: locate each video point of the target user in each sports event from the sports video of the target user in each sports event, and establish a suitable coordinate system in the video image of each video point. Usually, the lower left corner of the video screen can be taken as the origin, the horizontal rightward direction is the positive direction of the x-axis, and the vertical upward direction is the positive direction of the y-axis.

[0048] The key parts of each sport at each video point are extracted from the standard information of each sport in the gym. Based on the key parts, the coordinate values ​​of each key part are determined in the video image of each video point. Then, according to the general law of mass distribution of each part of the human body, the coordinates of each key part are weighted averaged. For example, the mass of the torso of the human body is relatively large, so it will be given a higher weight in the calculation, and the mass of the limbs is relatively small, so they will be given a lower weight. Then, the coordinates of the center of gravity of the body position are determined by this weighted average method.

[0049] Furthermore, the statistical standardization of the target user's body position in each sports event includes: W2-1 extracting the standard body position center of gravity position coordinates of each video point in each sports event from the standard information of each sports event in the fitness center, and using the distance formula to compare the standard body position center of gravity position coordinates of each video point in each sports event with the target user's body position center of gravity position coordinates of each video point in each sports event to obtain the deviation distance of the target user's body position center of gravity at each video point in each sports event, recorded as γ ij , j is the video point number, j=1,2,......,n.

[0050] W2-2. Statistical standardization of the target user’s body position in each sport δ i ′, γ i ′ is the permissible deviation distance of the center of gravity of the body position of the i-th sport, Δγi ′ is the deviation distance difference of the center of gravity of the body position of the i-th sports item for reference, and n is the number of sports items.

[0051] It should be added that γ i ′ and Δγ i ′ are extracted from the technical specification table of fitness center identification management.

[0052] W3, extract the muscle bulge ratio of each muscle bulge of the key parts of the target user in each sport and the action interval frequency of the target user in each sport from the sport video of the target user in each sport, and then calculate the action rhythm standardization degree δ of the target user in each sport i ″.

[0053] It should be added that the method of obtaining the muscle bulge ratio is as follows: when the muscle is not exerting force, the area of ​​the marked muscle area is calculated using the software function (such as the selection tool in Photoshop combined with related calculation functions), which is recorded as S0. For irregular shapes, it can be estimated by pixel statistics and other methods. When the muscle bulges, the area of ​​the marked muscle area is calculated again using the software function, which is recorded as S1, and then the muscle bulge ratio is calculated: Muscle bulge ratio: Thus, the bulge ratio of each muscle is calculated, and the average is calculated, and the calculation result is used as the muscle bulge ratio.

[0054] It should be added that the method of obtaining the action interval frequency of the target user in each sports item is as follows: starting from extracting the first key action frame from the sports video, observing the next identical key action frame appearing again, the frame interval between these two identical key action frames is the number of frames included in an action repetition cycle, for example, the number of frames from the frame in which the arms are bent to the lowest position for the first push-up to the frame in which the arms are bent to the lowest position for the next time.

[0055] If the frame rate of the video is f″ frames / second, then the time interval between each frame is

[0056] Assume that the number of frames in an action repetition cycle is σ, then the action repetition cycle time is

[0057]

[0058] The action interval frequency f is the number of times the action is repeated per unit time, using the formula For example, if the video frame rate is 30 frames per second and a push-up exercise repetition cycle contains 60 frames, then the action repetition cycle time is Action interval frequency

[0059]

[0060] Further, the statistical target user's action rhythm standardization in each sports event includes: W3-1, according to the muscle bulge ratio of each muscle bulge of the key parts of the target user in each sports event, the target user's muscle bulge qualification index λ in each sports event is statistically calculated. i .

[0061] Furthermore, the statistical target user's muscle bulge qualification index in each sports event includes: W3-1-1, calculating the average muscle bulge ratio of each muscle bulge of the target user's key parts in each sports event to obtain the average muscle bulge ratio of the target user's key parts in each sports event.

[0062] W3-1-2. Subtract the muscle bulge ratio of each muscle bulge of the key parts of the target user in each sports event from the average muscle bulge ratio to obtain the muscle bulge ratio difference of each muscle bulge of the key parts of the target user in each sports event.

[0063] W3-1-3. The muscle bulge ratio difference of each muscle bulge of the key parts of the target user in each sports event is introduced into the muscle bulge qualified index analysis model to obtain the muscle bulge qualified index of the target user in each sports event, which is recorded as λ i .

[0064] It should be added that the specific analysis process of the muscle bulge qualification index analysis model is as follows: if the muscle bulge ratio difference of a certain muscle bulge in a key part of a target user in a certain sports event is within the set reference muscle bulge ratio difference range, then the qualification index of the target user's muscle bulge in the key part of the sports event is recorded as 1.

[0065] If the muscle bulge ratio difference of a certain muscle bulge in a key part of the target user in a certain sports event is greater than the upper limit value of the reference muscle bulge ratio difference interval, then the ratio of the upper limit value of the reference muscle bulge ratio difference interval to the muscle bulge ratio difference will be used as the qualified index of the muscle bulge in the key part of the target user in the sports event, recorded as η1.

[0066] If the muscle bulge ratio difference of a certain muscle bulge in a key part of the target user in a certain sports event is less than the lower limit of the reference muscle bulge ratio difference interval, then the ratio of the muscle bulge ratio difference to the lower limit of the reference muscle bulge ratio difference interval is taken as the qualified index of the muscle bulge in the key part of the target user in the sports event, denoted as η2, and then the qualified index of each muscle bulge in the key part of the target user in the sports event is obtained.

[0067] The average calculation result of the qualified index of each muscle bulge of the key parts of the target user in the sports event is used as the qualified index of the muscle bulge of the target user in the sports event, and then the qualified index of the muscle bulge of the target user in each sports event is obtained.

[0068] W3-2. The target user's action interval frequency in each sport is recorded as f i .

[0069] W3-3. Extract the standard action interval frequency of each sport from the standard information of each sport in the gym, denoted as f i ′.

[0070] W3-4. Statistical deviation ratio θ of the target user’s action interval frequency in each sport i ,

[0071] W3-5. Count the target user's action interval frequency qualification index λ in each sports event i ′, θ′ is the action interval frequency deviation ratio of the setting reference.

[0072] It should be added that θ′ is extracted from the technical specification table of gym identification management.

[0073] In a specific embodiment, if the target user's exercise program is push-ups, the duration is 60 seconds.

[0074] By calculating the time interval between adjacent push-up exercise items, the repetition period t is obtained. For example, for the target user, the time interval between two consecutive push-up exercise items is measured to be 2.8 seconds, 3.0 seconds, 2.9 seconds, etc. (the average of multiple periods is taken to improve accuracy). The average repetition period is

[0075] According to the frequency formula Calculate the push-up interval frequency for the target user Converted into the number of times per minute, it is 0.34*60=20.4 times / minute.

[0076] The standard interval frequency of push-up exercise is f′=25 times / minute. The deviation ratio of the interval frequency of push-up exercise for the target user is calculated as

[0077]

[0078] The reference action interval frequency deviation ratio of the push-up sport is 0.1, and the target user's action interval frequency qualification index λ′ in the push-up sport is 0.

[0079] W3-6. Statistical analysis of target users’ action rhythm standardization in various sports i ″,δ i ″=λ i *u1+λ i ′*u2, u1 and u2 are the weights of the set muscle bulge qualification index and action interval frequency qualification index respectively, u1+u2=1, u2>u1.

[0080] It should be added that there is a correlation between movement interval frequency and injury risk. An unreasonable movement interval frequency often increases the risk of sports injuries. For example, in strength training programs, if the movement interval frequency is too fast, the muscles may not be able to fully contract and relax, thereby increasing the risk of muscle strain, joint injury, etc. Maintaining a standardized movement interval frequency can enable muscles and joints to work within a reasonable range of motion and reduce the possibility of injury. Although the muscle bulge qualification index is also related to the health of the muscles, in terms of preventing sports injuries, reasonable control of the movement interval frequency is more important. Therefore, u2>u1 is set. For the sake of analysis, u1 can be specifically taken as 0.4 and u2 can be specifically taken as 0.6.

[0081] W4. The mean calculation result of the joint standardization, position standardization and action rhythm standardization of the target user in each sports item is used as the sports behavior standardization of the target user in each sports item, denoted as

[0082] The user health identification module is used to identify the health information of the target user in each sports event, and the health information includes the exercise duration and interval duration.

[0083] It should be added that both the movement duration and the interval duration are located from the video images.

[0084] The user health analysis module is used to analyze the sports health standardization of the target user in each sports event based on the health information of the target user in each sports event.

[0085] See also Figure 3 As shown, for example, the analysis of the target user's sports health standardization in each sports event includes: Q1, according to the target user's sports time in each sports event, counting the target user's sports time standardization degree ψ of each sports event i .

[0086] Furthermore, the statistical standardization of the target user's exercise time in each sport includes: Q1-1, extracting the target user's permitted exercise time in each sport from a database.

[0087] Q1-2. Compare the target user's exercise time in each sport with his / her permitted exercise time. If the target user's exercise time in a certain sport is less than or equal to his / her permitted exercise time, the target user's exercise time standardization degree in that sport is recorded as 1.

[0088] Q1-3, if the target user's exercise time in a certain sport is longer than his permitted exercise time, the target user's exercise time in the sport and the permitted exercise time are recorded as and Then, the target user’s exercise duration standardization degree ξ in this sport is calculated. In this way, the target user's exercise duration standardization in each sports event is obtained and recorded as ψ i , ψ i The value is 1 or ξ.

[0089] Q2. According to the analysis method of the target user's exercise duration standardization in each sports event, the target user's exercise interval duration standardization in each sports event is obtained by similar analysis and recorded as ψ i ′.

[0090] Q3: Filter out the minimum value from the target user's exercise duration standardization and exercise interval duration standardization in each sports event as the target user's exercise health standardization in each sports event, denoted as

[0091] The embodiments of the present invention break the current lack of attention to the user behavior and health analysis dimensions by analyzing the target user's sports behavior norms and sports health norms in various sports events, thereby ensuring the user's sports health and reducing the probability of sports injuries in the user.

[0092] The user coach confirmation module is used to confirm the target user's required coach based on the target user's identity information, the target user's sports behavior norms in various sports events, and the sports health norms.

[0093] Exemplarily, confirming the target user's required coach includes: matching available coaches corresponding to the target user's identity information according to the target user's identity information.

[0094] It should be added that the matching process of matching the target user's identity information to the available coaches is as follows: when the target user's identity information is a new user, the available coaches corresponding to the junior membership level are used as the available coaches corresponding to the target user's identity information; otherwise, the target user's membership level is extracted from the target user's identity information, and then the available coaches of the target user's membership level are extracted from the gym management platform as the available coaches corresponding to the target user's identity information.

[0095] Extract the target user's sports behavior norms in each sports event and the target users’ sports health standards in various sports

[0096] Statistics on target users' sports standardization in various sports y1 and y2 are the weights of the reference sports behavior norms and sports health norms, y1+y2=1, y1>y2.

[0097] It should be added that correct sports behavior norms are the key to ensuring sports safety. If the sports behavior is not standardized, such as incorrect posture when lifting weights, it may cause serious injuries such as muscle strain, joint dislocation or even fracture. In contrast, although factors such as exercise duration and interval duration in sports health norms will also have an impact on the body, they generally will not immediately cause serious acute injuries like incorrect sports behaviors. Taking running as an example, incorrect running posture (such as excessive stride, excessive twisting of the body, etc.) may lead to accidents such as falls and sprained ankles during running, and these injuries may occur in a short time. Therefore, from the perspective of avoiding acute sports injuries, sports behavior norms are more important. Therefore, y1>y2 is set. For the convenience of analysis, y1 can be specifically taken as 0.6 and y2 can be specifically taken as 0.4.

[0098] from The sports item corresponding to the minimum value is selected as the target sports item, and then the fitness scores of the target sports items of the idle coaches corresponding to the target user's identity information are extracted from the fitness center management platform, so as to select the idle coach with the highest fitness score as the target user's required coach.

[0099] The embodiment of the present invention confirms the target user's required coach based on the target user's identity information, the target user's exercise behavior norms in various sports events, and the exercise health norms, thereby avoiding the shortcomings of the current coaching confirmation based only on the user's identity level, ensuring the user's training safety, thereby improving the user's training effect, and at the same time optimizing the allocation of coaching resources, thereby improving the coach's work efficiency.

[0100] The database is used to store the user's permitted exercise time for each sport, store user information of a fitness center, and store standard information of each sport in the fitness center.

[0101] It should be added that the permitted exercise time of the user in each sport event refers to the maximum value of each historical exercise time of the user in each sport event as the permitted exercise time of the user in each sport event.

[0102] It should be added that the standard information of each sport in the gym includes: standard joint angle intervals of each key action in each sport, standard body center of gravity position coordinates of each video point in each sport, and standard action interval frequency of each sport.

[0103] The gym management terminal is used to train target users according to their needs and send instructions to the gym management platform.

[0104] The embodiments of the present invention avoid the current problem of insufficient attention to user exercise behavior by identifying and analyzing the user's identity, behavior and health, and realize personalized services for users, thereby improving the targeted guidance to users, while ensuring the timeliness of user behavior processing, improving the user's usage experience, and thus improving the user's retention rate.

[0105] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. An AI-based intelligent identification and management system for a fitness center, characterized by: The system includes: User identification module, used to identify the face image of the target user; A user identity confirmation module is used to confirm the identity information of the target user based on the face image of the target user; The user behavior recognition module is used to track and shoot the target user's movement process according to the target user's identity information, and then obtain the target user's movement video in various sports events; The user behavior analysis module is used to analyze the standardization of the target user's sports behavior in each sports event based on the target user's sports video in each sports event; The user health identification module is used to identify the target user's health information in each sports event, including exercise duration and interval duration; The user health analysis module is used to analyze the target user's sports health standardization in each sports event based on the target user's health information in each sports event; The user coach confirmation module is used to confirm the target user's required coach based on the target user's identity information, the target user's sports behavior norms in various sports events, and the sports health norms; A database for storing the permitted exercise time of users in various sports, storing user information of fitness centers, and storing standard information of various sports in fitness centers; The gym management terminal is used to train the target users according to their needs and send instructions to the gym management platform.

2. According to claim 1, the AI-based intelligent identification and management system for a fitness center is characterized by: The confirmation of the identity information of the target user includes: Extracting stored facial images of each user in the gym from the gym management platform; The target user's facial recognition image is matched and compared with the stored facial images of each user in the gym through AI technology; When the face recognition image of the target user is consistent with the face image stored in the gym, the identity information of the user is extracted from the user information of the gym as the identity information of the target user; When the face recognition image of the target user is inconsistent with the stored face images of each user in the gym, the new user is used as the identity information of the target user.

3. The AI-based intelligent identification and management system for a fitness center according to claim 1, characterized in that: The analysis of the target user's sports behavior standardization in each sports event includes: W1, extract the joint angles and joint angle change rates of each key action of the target user in each sport from the target user's sports video in each sport, and then count the joint standardization degree δ of the target user in each sport i , i is the sports item number, i=1,2,......,m; W2, extract the coordinates of the target user's body position center of gravity at each video point in each sport from the target user's sports video in each sport, and then count the target user's body position standardization degree δ′ in each sport i ; W3, extract the muscle bulge ratio of each muscle bulge of the key parts of the target user in each sport and the action interval frequency of the target user in each sport from the sport video of the target user in each sport, and then calculate the action rhythm standardization degree δ of the target user in each sport i ″; W4. The mean calculation result of the joint standardization, position standardization and action rhythm standardization of the target user in each sports item is used as the sports behavior standardization of the target user in each sports item, denoted as 4. The AI-based fitness center intelligent identification management system according to claim 3 is characterized by: The statistical target user's joint standardization in each sport includes: Extract standard information of various sports in the gym from the database, and then extract standard joint angle intervals of various key movements in each sports; If the joint angle of a key action of a user in a certain sport is within the standard joint angle range, the key action of the user in the sport is recorded as a qualified key action, and the number of qualified key actions of the sport is counted; The number of key actions of the sport is counted, and the ratio of the number of qualified key actions to the number of key actions is used as the qualified ratio of joint angles of the sport. In this way, the qualified ratio of joint angles of the target user in each sport is obtained, which is recorded as τ i ; According to τ i The statistical method is similar to the statistical method to obtain the qualified ratio of the target user's joint angle change rate in each sports event, which is recorded as τ′ i ; τ i and τ′ i The mean value is calculated and the mean value is used as the joint standardization degree of the target user in each sports event, which is recorded as δ i .

5. The AI-based intelligent identification and management system for a fitness center according to claim 3 is characterized by: The statistical target user's body position standardization in each sport event includes: The standard body position center coordinates of each video point in each sport are extracted from the standard information of each sport in the gym. The deviation distance of the body position center coordinates of each video point in each sport is obtained by using the distance formula with the standard body position center coordinates of each video point in each sport, which is recorded as γ ij , j is the video point number, j = 1, 2, ..., n; Statistical target user's body position standardization δ′ in each sport i , γ′ i is the permissible deviation distance of the center of gravity of the body position for the i-th sport, Δγ′ i is the deviation distance difference of the center of gravity of the body position of the i-th sport for reference, and n is the number of sports.

6. The AI-based intelligent identification and management system for a fitness center according to claim 3, characterized in that: The statistical target user's action rhythm standardization in each sports event includes: According to the muscle bulge ratio of each key part of the target user in each sport, the target user's muscle bulge qualification index λ in each sport is calculated i ; The target user's action interval frequency in each sports item is recorded as f i ; Extract the standard action interval frequency of each sport from the standard information of each sport in the gym, denoted as f′ i ; Statistical target user's action interval frequency deviation ratio θ in each sports event i , Calculate the target user's action interval frequency qualification index λ′ in each sports event i , θ′ is the action interval frequency deviation ratio of the setting reference; Statistical target user's action rhythm standardization degree δ″ in each sports event i ,δ″ i =λ i *u1+λ′ i *u2, u1 and u2 are the weights of the set muscle bulge qualification index and action interval frequency qualification index respectively, u1+u2=1, u2>u1.

7. The AI-based intelligent identification and management system for a fitness center according to claim 6, characterized in that: The statistical target user's muscle bulge qualification index in each sport includes: The muscle bulge ratios of each muscle bulge of the key parts of the target user in each sports event are averaged to obtain the average muscle bulge ratio of the key parts of the target user in each sports event; Subtract the muscle bulge ratio of each muscle bulge of the key parts of the target user in each sport from the average muscle bulge ratio to obtain the muscle bulge ratio difference of each muscle bulge of the key parts of the target user in each sport; The muscle bulge ratio difference of each muscle bulge of the key parts of the target user in each sports event is introduced into the muscle bulge qualified index analysis model to obtain the muscle bulge qualified index of the target user in each sports event, which is recorded as λ i .

8. The AI-based intelligent identification and management system for a fitness center according to claim 1, characterized in that: The analysis of the target user's sports health standardization in each sports event includes: Q1. According to the target user's exercise time in each sport, calculate the target user's exercise time standardization degree ψ in each sport i ; Q2. According to the analysis method of the target user's exercise duration standardization in each sports event, the target user's exercise interval duration standardization in each sports event is obtained by similar analysis and recorded as ψ′ i ; Q3: Filter out the minimum value from the target user's exercise duration standardization and exercise interval duration standardization in each sports event as the target user's exercise health standardization in each sports event, denoted as 9. The AI-based intelligent identification and management system for a fitness center according to claim 1, characterized in that: The statistical target user's exercise duration standardization in each sport event includes: Extract the target user's permitted exercise time in each sport from the database; Compare the target user's exercise time in each sport with his / her permitted exercise time. If the target user's exercise time in a sport is less than or equal to his / her permitted exercise time, the target user's exercise time standardization degree in the sport is recorded as 1. If the target user's exercise time in a certain sport is longer than his permitted exercise time, the target user's exercise time in the sport and the permitted exercise time are recorded as and Then, the target user’s exercise duration standardization degree ξ in this sport is calculated. In this way, the target user's exercise duration standardization in each sports event is obtained and recorded as ψ i , ψ i The value is 1 or ξ.

10. The AI-based intelligent identification and management system for a fitness center according to claim 3, characterized in that: The coach who confirms the needs of the target users includes: According to the identity information of the target user, matching each available coach corresponding to the identity information of the target user; Extract the target user's sports behavior norms in each sports event and the target users’ sports health standards in various sports Count the target user's sports standardization in each sports event i , y1 and y2 are the weights of the reference sports behavior norms and sports health norms, y1+y2=1, y1>y2. From l i The sports item corresponding to the minimum value is selected as the target sports item, and then the fitness scores of the target sports items of the idle coaches corresponding to the target user's identity information are extracted from the fitness center management platform, so as to select the idle coach with the highest fitness score as the target user's required coach.

Images