A virtual assistant method and system for a gym management system

Abstract

The present invention discloses a virtual assistant method and system for a gym management system, comprising: collecting an image of a target human body, cutting the image of the target human body according to preset human body parts to obtain a human body block image; determining a symmetry point, a symmetry axis or a symmetry plane according to the collected human body block image and a fitness exercise type, and performing a symmetry analysis of the cut human body block image according to the marked symmetry point, symmetry axis or symmetry plane; judging the balance of the target human body in a corresponding fitness exercise type according to a symmetry analysis parameter result of the human body block image, and if the symmetry analysis result of the human body block image of the target human body does not meet a preset standard parameter range, generating a motion correction strategy for the human body block of the target human body in the corresponding fitness exercise type; and sending the motion correction strategy to a fitness manager, who guides the correction of the fitness exercise of the target human body according to the motion correction strategy.

Description

Technical Field

[0001] The present invention relates to the field of management technology, and in particular to a virtual assistant method and system for a gym management system. Background Art

[0002] At present, the main functions of the gym management system are still mainly based on the members' recharge accounting, course reservation and sports information recording functions. For example, the existing gym management system can provide all the courses that can be booked, including course name, time, coach, location, etc., and make online course reservations through the gym management system; or the existing gym management system provides the function of recording members' health indicators, such as members' weight, body fat percentage, etc., and provides professional fitness suggestions based on members' health indicators. However, the above-mentioned existing gym management system has the following technical problems: it mainly provides a relatively static data recording function, and professional coaches need to provide on-site guidance for problems in the fitness process. Existing fitness coaches often provide guidance to a small number of customers throughout the fitness process, which leads to high human cost for fitness coaches. (Human body division, symmetry analysis) (Synchronous timing symmetry analysis, asynchronous timing symmetry analysis). Summary of the invention

[0003] One of the inventive purposes of the present invention is to provide a virtual assistant method and system for a gym management system, wherein the method and system utilize a target recognition model to identify a target human body being trained, and divide the target human body into motion blocks according to the type of fitness training, and obtain whether the current fitness training posture meets the training standards by performing on-site real-time analysis of the divided motion blocks, and automatically send the analysis results to gym managers such as fitness coaches, thereby effectively improving the management efficiency of the fitness process and effectively reducing the management costs of the gym through auxiliary management.

[0004] Another inventive object of the present invention is to provide a virtual assistant method and system for a gym management system, wherein the method and system use human motion block images collected and divided by a camera to perform symmetry analysis according to the type of fitness exercise, and the symmetry analysis can provide a balance warning for fitness exercise, wherein the symmetry analysis includes up-and-down and front-and-back symmetry analysis, and the symmetry analysis can be used to determine the degree to which the target human body has mastered the posture of the exercise equipment or its own movement posture during exercise, thereby reducing accidental injuries during exercise caused by unbalanced posture control of the exercise equipment or unbalanced movement posture, and effectively improving the safety of fitness exercise.

[0005] Another inventive object of the present invention is to provide a virtual assistant method and system for a gym management system, which utilizes posture recognition of different parts of the human body to identify balance, without the need for external wearable devices or force sensors to perform strength detection, and can have better exercise process management efficiency and effect on the basis of reducing hardware equipment detection. Moreover, balance detection based on human body part recognition has better real-time and accuracy than the management of the coach itself, which is conducive to assisting fitness coaches in managing the fitness process.

[0006] In order to achieve at least one of the above-mentioned invention objects, the present invention further provides a virtual assistant method for a gym management system, the method comprising:

[0007] Collecting an image of a target human body, cutting the target human body into images according to preset human body parts, and obtaining human body block images;

[0008] Marking and determining symmetry points, symmetry axes or symmetry planes according to the collected human body block images and fitness exercise types, and performing symmetry analysis on the cut human body block images according to the marked symmetry points, symmetry axes or symmetry planes;

[0009] According to the symmetry analysis parameter result of the human body block image, the balance of the target human body in the corresponding fitness exercise type is judged; if the symmetry analysis result of the human body block image of the target human body does not meet the preset standard parameter range, a motion correction strategy for the human body block of the corresponding target human body in the corresponding fitness exercise type is generated;

[0010] The movement correction strategy is sent to a fitness manager, who guides the correction of the fitness movement of the target person according to the movement correction strategy.

[0011] According to one of the preferred embodiments of the present invention, the method for obtaining the human body block image by image cutting includes: after acquiring the image of the target human body, using a contour recognition algorithm to segment the target human body into a head and neck area image, a torso area image, a left hand area image, a right hand area image, a left leg area image and a right hand area image, wherein the left hand area image and the right hand area image are used as binding images for symmetry analysis of corresponding fitness exercise types; and the left leg area image and the right leg area image are used as binding images for symmetry analysis of corresponding fitness exercise types.

[0012] According to another preferred embodiment of the present invention, the symmetry analysis includes symmetry analysis of synchronous time series and symmetry analysis of asynchronous time series; wherein the symmetry analysis of the synchronous time series includes: obtaining a human body block image after cutting the target human body image during the fitness process, automatically assigning the detected human body block image for symmetry analysis according to the fitness exercise type of the target human body, and automatically determining the symmetry point, symmetry axis or symmetry plane, obtaining the time series of each assigned human body block image, selecting the pixel distribution of each assigned human body block image in at least one frame of the same time series relative to the determined symmetry point, symmetry axis or symmetry plane, and judging the symmetry parameter value of the assigned human body block image in the current same time series according to the pixel distribution.

[0013] According to another preferred embodiment of the present invention, the method for symmetry analysis of asynchronous time series includes: automatically generating the time interval length of the asynchronous time series according to the rhythm of the fitness exercise type of the target human body, and automatically allocating the detected human body block images for symmetry analysis, and automatically determining the symmetry point, symmetry axis or symmetry plane, and obtaining the pixel distribution of the human body block images allocated and detected in at least two frames in different time series relative to the symmetry point, symmetry axis or symmetry plane according to the time interval length of the asynchronous time series, and judging the symmetry parameter values ​​of the allocated human body block images in the current at least two frames of different time series according to the pixel distribution.

[0014] According to another preferred embodiment of the present invention, the method for calculating the symmetry parameter value includes: constructing a coordinate system of an image acquisition interface, wherein the coordinate system of the image acquisition interface constructs a corresponding coordinate system origin, coordinate axis or coordinate system plane with the determined symmetry point, symmetry axis or symmetry plane, and simultaneously obtaining pixel coordinate values ​​of the human body block images assigned for detection in the same time series or different time series according to the coordinate system origin, coordinate axis or coordinate system plane, obtaining pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the pixel coordinate values, summing and calculating the pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the coordinate values ​​in the direction perpendicular to the symmetric coordinate axis, obtaining the summation result of the pixel average coordinate values ​​of different human body block images, pre-setting a symmetry parameter range, and when the summation result is within the symmetry parameter range, the corresponding human body block of the current target human body meets the symmetry requirements in the corresponding fitness exercise type.

[0015] According to another preferred embodiment of the present invention, the method for calculating the symmetry parameter value includes: constructing a coordinate system of an image acquisition interface, wherein the coordinate system of the image acquisition interface constructs a corresponding coordinate system origin, coordinate axis or coordinate system plane with the determined symmetry point, symmetry axis or symmetry plane, and simultaneously obtains pixel coordinate values ​​of the human body block images assigned for detection in the same time series or different time series according to the coordinate system origin, coordinate axis or coordinate system plane, obtains pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the pixel coordinate values, and calculates distance values ​​of the pixel average coordinate values ​​of different human body block images used for symmetry analysis relative to the symmetry point, symmetry axis or symmetry plane, and pre-sets a symmetry parameter range with a distance value range. When the distance value is within the symmetry range, the corresponding human body block of the current target human body meets the symmetry requirements in the corresponding fitness exercise type.

[0016] According to another preferred embodiment of the present invention, the virtual assistant also includes a consistency analysis, wherein the consistency analysis method includes: selecting the coordinate values ​​of the pixel average coordinate values ​​of the different human body block images used for symmetry analysis in the direction of the consistency coordinate axis, and calculating the difference in the coordinate values ​​of the pixel average coordinate values ​​of the different human body block images in the direction of the consistency coordinate axis, and presetting a consistency parameter range. When the difference in the coordinate values ​​of the pixel average coordinate values ​​of the different human body block images in the direction of the consistency coordinate axis is within the consistency parameter range, the consistency analysis of the current target human body in the corresponding fitness exercise type meets the requirements.

[0017] According to another preferred embodiment of the present invention, when the symmetry analysis result and / or consistency analysis result of the target human body in the corresponding fitness exercise type does not meet the requirements, the parameter values ​​of the symmetry analysis results and / or consistency analysis results of the target human body different body block images and the difference between the upper and lower limits of the target parameter range are automatically calculated, and the parameter values ​​of the symmetry analysis results and / or consistency analysis results and the difference between the upper and lower limits of the target parameter range are converted into a motion correction strategy and sent to the gym manager.

[0018] In order to achieve at least one of the above-mentioned invention objectives, the present invention further provides a virtual assistant system for a gym management system, and the system executes the above-mentioned virtual assistant method for a gym management system.

[0019] The present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the virtual assistant method for a gym management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1Shown is a flow chart of a virtual assistant method for a gym management system according to the present invention. DETAILED DESCRIPTION

[0021] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0022] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0023] Please combine Figure 1The present invention discloses a virtual assistant method and system for a gym management system, the method comprising the following steps: first, a camera needs to be fixedly configured on fitness equipment or a position, the camera may include but is not limited to a 3D camera, and the 3D camera may be used to collect plane information and depth information including an image. The image information collected by the 3D camera is input into a trained target detection model for human image detection, wherein the target detection model may include but is not limited to an SSD (Single Shot MultiBox Detector) model and a Faster R-CNN (Regions with Convolutional Neural Networks) model, etc., and the image collected by the camera is subjected to human recognition through the trained target detection model to obtain a target human image. The target detection models exemplified above are all prior art, and the present invention does not improve the target detection model itself, so the present invention does not explain in detail the implementation method of the target detection model exemplified above. After obtaining the target human body image, the target human body image is further cut according to the human body block contour, wherein the present invention cuts the target human body image into the following parts according to the contour: head and neck area image, torso area image, left hand area image, right hand area image, left leg area image and right hand area image and other human body block images. The human body block image can be cut according to the contour structure of the human body part, and can adopt contour area-based segmentation algorithms including but not limited to K-means algorithm and Sobel algorithm. The above algorithms are all prior art, and the present invention does not improve the above existing algorithms, so the present invention will not be described in detail. The present invention performs symmetry analysis of human body block images in the same frame in the same time sequence according to the type of fitness exercise for different human body block images, and can also perform symmetry analysis of human body block images in different frames in different time sequences according to the type of fitness exercise. In one of the preferred embodiments of the present invention, the present invention can also perform consistency analysis on human body block images, so that the human body block images obtained for detection include symmetry analysis results and / or consistency analysis results, and further generate corresponding fitness exercise correction strategies based on the symmetry analysis results and consistency analysis results, and send the fitness exercise correction strategies to the corresponding fitness administrators, who can provide on-site guidance based on the fitness exercise correction strategies. Therefore, the method and system of the present invention have a good auxiliary effect in the fitness process, and effectively reduce the manpower cost of on-site fitness guidance.

[0024] Specifically, since the posture forms required to be achieved by different types of fitness exercises are different, different posture forms can be disassembled into a balanced posture and a consistent posture, wherein the balanced posture can be obtained by symmetry analysis of different human body blocks of the target human body, and the consistent posture can be obtained by consistency analysis of different human body blocks of the target human body. It should be noted that the symmetry analysis of the exercise posture can effectively explain the force balance of the target human body during the exercise process. As is well known, when the posture imbalance occurs during the fitness exercise, it is very easy to cause muscle strain, cartilage damage, dislocation, and fitness equipment from being detached and smashed. For example, for fitness equipment with a pull bar, it is necessary to keep the left and right palms accurately holding the pull rods of the horizontal bar, and keep the relative left and right balance to achieve a safe and effective fitness exercise effect. In one of the preferred embodiments of the present invention, the symmetry analysis also includes fitness exercises that need to rotate around a fixed point. With respect to the consistent posture, the present invention needs to analyze whether different body blocks of the target human body remain relatively consistent in at least one direction of the up and down direction, the left and right direction, or the front and back direction of the image. By analyzing the consistency of different body blocks of the target human body, the fitness training standards can be improved, thereby improving the fitness training effect.

[0025] It is worth mentioning that the symmetry analysis and consistency analysis of the human body blocks corresponding to the target human body in the present invention also include the symmetry analysis of the target human body blocks of the synchronous time sequence, the symmetry analysis of the target human body blocks of the asynchronous time sequence, the consistency analysis of the target human body blocks of the synchronous time sequence, and the consistency analysis of the target human body blocks of the asynchronous time sequence. The symmetry analysis and consistency analysis of the asynchronous time sequence in the present invention are to solve the problem of maintaining symmetry or consistency at the rhythm points of different time sequences in the exercise movements at different exercise rhythms. At this time, the target human body can effectively analyze the maintenance status of the fitness exercise rhythm posture, and improve the analysis effect of the fitness exercise rhythm posture.

[0026] The method for analyzing the symmetry of the human body block corresponding to the target human body comprises: obtaining the target human body image through the camera and dividing the target human body into head and neck area image, trunk area image, left hand area image, right hand area image, left leg area image and right hand area image and other human body block images. Automatically generate symmetry points, symmetry axes or symmetry planes according to the type of fitness exercise selected by the target human body. The symmetry points, symmetry axes or symmetry planes can be constructed according to the position of the corresponding fitness equipment or fixed pixel points can be selected on the image plane, and the pixel columns are used as the symmetry points, symmetry axes or symmetry planes of the human body block corresponding to the target human body; or in another preferred embodiment of the present invention, the center point of the target human body can be automatically identified by the target detection model, and the human body symmetric pixel columns are used as the symmetry axes or symmetry planes, wherein the symmetry planes can be obtained by the depth information of the 3D panoramic camera, and the symmetry planes can be mapped to a two-dimensional plane as a symmetry axis description. The present invention does not specifically limit the generation method of the symmetry points, symmetry axes or symmetry planes.

[0027] After obtaining the pixel information of the symmetric point, symmetric axis or symmetric plane in the image, a coordinate system is further constructed based on the symmetric point, symmetric axis or symmetric plane, wherein the coordinate system may be a two-dimensional plane coordinate system. If there is a three-dimensional image of a three-dimensional panoramic camera, a three-dimensional coordinate system may be constructed. The present invention is illustrated by taking a two-dimensional plane coordinate system as an example: in the present invention, the central pixel point of the image captured by the camera is taken as the coordinate origin O, the pixel column perpendicular to the upper and lower image boundaries and passing through the coordinate origin is taken as the Y-axis, the pixel column perpendicular to the left and right image boundaries and passing through the coordinate origin is taken as the X-axis, and the coordinate value increment of a pixel is 1. Define a human body block image i, and another human body block image used for symmetry analysis is j, define P as a pixel distribution set, p represents the corresponding individual pixel coordinate value, and t represents the time series. Then, in the symmetry analysis of the synchronous time series, it is necessary to calculate the pixel distribution set P of the human body block image i and another human body block image j in the same frame under the time series t. i,t =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ] and P j,t =[p j,1 ,p j,2 ,p j,3 ,p j,4 ....p j,m ] on the same coordinate axis and , and calculate the average value on the same coordinate axis and The coordinate values ​​perpendicular to the symmetry axis are summed. For example, in the present invention, the coordinate values ​​perpendicular to the symmetry axis are the coordinate values ​​on the x-axis. Then, the symmetry analysis parameter f(x) of the human body block image based on the sum of the x-axis coordinate values ​​can be obtained. t ). The symmetry parameter range λ is preset. If the symmetry parameter range λ∈[-20,20], then when the symmetry analysis parameter f(x t )∈[-20,20], the current symmetry analysis is passed, and it is determined that the current target human body blocks i and j meet the symmetry analysis requirements under the same time series t, that is, they meet the balance requirements.

[0028] In one of the preferred embodiments of the present invention, the symmetry analysis can also calculate the pixel distribution set P of the human body block image i and another human body block image j in the same frame in the time series t by making the pixel distribution set P relative to the difference in distance between the symmetry points, symmetry axes or symmetry planes. i,t =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ] and P j,t =[p j,1 ,p j,2 ,p j,3 ,p j,4 ....p j,m ] on the same coordinate axis and After that, calculate the distance of the coordinate perpendicular to the axis of symmetry corresponding to the average value relative to the axis of symmetry. If the coordinate values ​​are x1 and x2 respectively, then it is necessary to calculate the distance of x1 and x2 relative to the y-axis of the axis of symmetry, and calculate the absolute value of the difference between the two distances. , when the absolute value of the difference ∈[-20,20], the current symmetry analysis passes.

[0029] Of course, in another preferred embodiment of the present invention, the symmetry analysis method for the symmetric point includes: calculating the pixel distribution set P of the human body block image i and another human body block image j in the same frame in the time series t i,t =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ] and P j,t =[p j,1 ,p j,2 ,p j,3 ,p j,4 ....pj,m ] on the same coordinate axis and Then, obtain the specific coordinates of the average value =(x1,y1) and =(x2,y2), then calculate the distance between the two coordinates relative to the symmetric point O=(0,0) and , and calculate the absolute value of the difference between distances l1 and l2 , when the difference ∈[-20,20], the current symmetry analysis passes. It should be noted that the distance between symmetric points as the symmetry analysis method needs to be set according to the type of fitness exercise.

[0030] In another preferred embodiment of the present invention, the asynchronous sequence symmetry analysis includes: automatically generating a time interval length s of the asynchronous time series according to the rhythm of the fitness exercise type of the target human body, and automatically allocating the detected human body block images for symmetry analysis according to the target detection model and the image cutting model, and automatically determining the symmetry point, symmetry axis or symmetry plane by including but not limited to the pixel center point of the camera-captured image and the pixel center point of the human body target, and obtaining the pixel distribution of at least two frames of the human body block images allocated and detected in different time series relative to the symmetry point, symmetry axis or symmetry plane according to the time interval length of the asynchronous time series, and judging the symmetry parameter values ​​of the currently allocated human body block images of at least two frames of different time series according to the pixel distribution.

[0031] For example: Calculate the pixel distribution set P of the human body block image i in the next frame of the time series t i,t =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ], and calculate the pixel distribution set P of another human body block image j corresponding to different frames in another time series t+s according to the time interval length s of the asynchronous time series j,t+s =[p j,1 ,p j,2 ,p j,3 ,p j, 4....p j,m ], and further calculate the coordinate average of the two human body block images i and j of the asynchronous time series on the same coordinate axis and After obtaining the coordinate average value, the symmetry analysis can be performed in the same way as the above-mentioned synchronous time series. The symmetry analysis method has been illustrated above, and the present invention will not be described in detail.

[0032] Furthermore, the present invention also requires a consistency analysis of the human body block image i and the human body block image j, wherein the consistency analysis includes: obtaining a pixel distribution set P of the human body block image i corresponding to the video frame at the time sequence t i,t =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ], and calculate the pixel distribution set P of the human body block image i corresponding to the video frame at time sequence t i,t The average value of coordinates on the same axis At the same time, the pixel distribution set P of the human body block image i corresponding to the video frame at time series t+s is calculated according to the fitness exercise rhythm point i,t+s =[p i,1 ,p i,2 ,p i,3 ,p i,4 ....p i,n ], and calculate the average coordinates of the human body block image i on the same coordinate axis at the time series t+s , and calculate the average and The absolute value of the difference , when the absolute value of the difference When the preset consistency parameter range β is met, the consistency analysis of the human body block image i in different time series meets the requirements. Similarly to the human body block image j, in another preferred embodiment of the present invention, the consistency analysis of the human body block image can also be applied to different human body block images.

[0033] In one of the preferred embodiments of the present invention, when the symmetry analysis result and / or consistency analysis result of the target human body in the corresponding fitness exercise type does not meet the requirements, the parameter values ​​of the symmetry analysis results and / or consistency analysis results of different body block images of the target human body and the difference between the upper and lower limits of the target parameter range are automatically calculated, and the parameter values ​​of the symmetry analysis results and / or consistency analysis results and the difference between the upper and lower limits of the target parameter range are converted into a motion correction strategy and sent to the gym manager, wherein the motion correction strategy may include but is not limited to the action posture of raising or lowering a body block corresponding to a certain body block image, etc.

[0034] The embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wire segments, 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 above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, electrical wire, optical cable, RF, etc., or any suitable combination of the foregoing.

[0035] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0036] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the principles, the implementation methods of the present invention may be deformed or modified in any way.

Claims

1. A virtual assistant method for a gym management system, characterized in that: The method comprises: Collecting an image of a target human body, cutting the target human body into images according to preset human body parts, and obtaining human body block images; Marking and determining symmetry points, symmetry axes or symmetry planes according to the collected human body block images and fitness exercise types, and performing symmetry analysis on the cut human body block images according to the symmetry points, symmetry axes or symmetry planes determined by the marking; The symmetry of the target human body under the corresponding fitness exercise type is judged according to the symmetry analysis parameter result of the human body block image, and if the symmetry analysis result of the human body block image of the target human body does not meet the preset standard parameter range, a motion correction strategy for the human body block of the corresponding target human body under the corresponding fitness exercise type is generated; Sending the motion correction strategy to a fitness manager, who instructs the target person to correct his / her fitness exercise according to the motion correction strategy; The symmetry analysis includes symmetry analysis of synchronous time series and symmetry analysis of asynchronous time series; wherein the symmetry analysis of the synchronous time series includes: obtaining a human body block image after the target human body image is cut during the fitness process, automatically assigning the detected human body block image for symmetry analysis according to the fitness exercise type of the target human body, and automatically determining the symmetry point, symmetry axis or symmetry plane, obtaining the time series of each assigned human body block image, selecting the pixel distribution of each assigned human body block image in at least one frame of the same time series relative to the determined symmetry point, symmetry axis or symmetry plane, and judging the symmetry parameter value of the assigned human body block image in the current same time series according to the pixel distribution; The method for symmetry analysis of asynchronous time series includes: automatically generating the time interval length of the asynchronous time series according to the rhythm of the fitness exercise type of the target human body, automatically allocating the detected human body block images for symmetry analysis, and automatically determining the symmetry point, symmetry axis or symmetry plane, respectively obtaining the pixel distribution of at least two frames of human body block images allocated and detected in different time series relative to the symmetry point, symmetry axis or symmetry plane according to the time interval length of the asynchronous time series, and judging the symmetry parameter values ​​of the allocated human body block images of at least two frames of different time series according to the pixel distribution.

2. A virtual assistant method for a gym management system according to claim 1, characterized in that: The method for obtaining the human body block image by image cutting comprises: after acquiring the image of the target human body, using a contour recognition algorithm to segment the target human body into a head and neck area image, a trunk area image, a left hand area image, a right hand area image, a left leg area image and a right leg area image, wherein the left hand area image and the right hand area image are used as binding images for symmetry analysis of corresponding fitness exercise types; and the left leg area image and the right leg area image are used as binding images for symmetry analysis of corresponding fitness exercise types.

3. The virtual assistant method for a gym management system according to claim 1, characterized in that: The method for calculating the symmetry parameter value includes: constructing a coordinate system of an image acquisition interface, wherein the coordinate system of the image acquisition interface constructs a corresponding coordinate system origin, coordinate axis or coordinate system plane with the determined symmetry point, symmetry axis or symmetry plane, and simultaneously obtaining pixel coordinate values ​​of the human body block images assigned for detection in the same time series or different time series according to the coordinate system origin, coordinate axis or coordinate system plane, obtaining pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the pixel coordinate values, summing and calculating the pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the coordinate values ​​in the direction perpendicular to the symmetry coordinate axis, obtaining the summation result of the pixel average coordinate values ​​of different human body block images, presetting a symmetry parameter range, and when the summation result is within the symmetry parameter range, the corresponding human body block of the current target human body meets the symmetry requirement in the corresponding fitness exercise type.

4. The virtual assistant method for a gym management system according to claim 1, characterized in that: The method for calculating the symmetry parameter value includes: constructing a coordinate system of an image acquisition interface, wherein the coordinate system of the image acquisition interface constructs a corresponding coordinate system origin, coordinate axis or coordinate system plane with the determined symmetry point, symmetry axis or symmetry plane, and simultaneously obtaining pixel coordinate values ​​of the human body block images assigned for detection in the same time series or different time series according to the coordinate system origin, coordinate axis or coordinate system plane, obtaining pixel average coordinate values ​​of different human body block images used for symmetry analysis according to the pixel coordinate values, and calculating the distance value of the pixel average coordinate value of the different human body block images used for symmetry analysis relative to the symmetry point, symmetry axis or symmetry plane, and a preset symmetry parameter range. When the distance value is within the symmetry parameter range, the corresponding human body block of the current target human body meets the symmetry requirement in the corresponding fitness exercise type.

5. The virtual assistant method for a gym management system according to claim 3, characterized in that: The virtual assistant also includes consistency analysis, wherein the consistency analysis method includes: selecting the coordinate values ​​of the pixel average coordinate values ​​of different human body block images used for symmetry analysis in the direction of the consistency coordinate axis, and calculating the difference in the coordinate values ​​of the pixel average coordinate values ​​of the same human body block images in different time series in the direction of the consistency coordinate axis, and presetting a consistency parameter range. When the difference in the coordinate values ​​of the pixel average coordinate values ​​of the same human body block image in the direction of the consistency coordinate axis is within the consistency parameter range, the consistency analysis of the current target human body in the corresponding fitness exercise type meets the requirements.

6. A virtual assistant method for a gym management system according to claim 5, characterized in that: When the symmetry analysis results and / or consistency analysis results of the target human body in the corresponding fitness exercise type do not meet the requirements, the parameter values ​​of the symmetry analysis results and / or consistency analysis results of different body block images of the target human body and the difference between the upper and lower limits of the target parameter range are automatically calculated, and the parameter values ​​of the symmetry analysis results and / or consistency analysis results and the difference between the upper and lower limits of the target parameter range are converted into a motion correction strategy and sent to the fitness manager.

7. A virtual assistant system for a gym management system, characterized in that: The system executes a virtual assistant method for a gym management system as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the virtual assistant method for a gym management system as described in any one of claims 1 to 6.

Images