Method and device for judging volleyball passing pad ball training based on visual technology

Abstract

The application discloses a method and device for judging volleyball passing and padding ball training based on visual technology and relates to the technical field of intelligent sports.The method comprises the following steps: a deep learning model detection process, real-time frame detection of a target frame and finger key points; a volleyball passing and padding ball judgment process, judgment of whether passing and padding ball is successful through the relationship between key point coordinates and target frame coordinates; and a passing and padding ball logic judgment process, accurate counting of the number of times of successful passing and padding ball through the setting of a sports state flag and passing and padding ball judgment logic.The method and device for judging volleyball passing and padding ball training based on visual technology provided by the application realize real-time and accurate counting based on computer visual technology in combination with the state control of passing and padding ball, and the fairness and accuracy of examination results are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart sports, in particular to a method and device for judging volleyball passing and padding ball training based on visual technology. BACKGROUND

[0002] Passing and padding ball is one of the key techniques in volleyball, and accurate judgment of the result of passing and padding ball can help athletes and coaches better understand and improve their techniques, and improve the performance and competitiveness of athletes. Passing and padding ball movement requires rapid movement, flexible turning and accurate passing of the ball, which can improve the body coordination, sensitivity and reaction speed of young people. Passing and padding ball movement requires young people to cooperate with their teammates, which can cultivate their team cooperation spirit and communication ability. Through passing and padding ball movement, young people can improve their self-confidence and self-esteem, and enhance their competitive consciousness and challenge spirit.

[0003] Due to the complexity of the judgment and counting rules of passing and padding ball, the current volleyball passing and padding ball training generally adopts the method of manual counting, which requires long-time concentration of attention, and the work intensity of the referee is high, which is easy to cause fatigue and inattention, and the referee is prone to misjudgment, missed counting and wrong counting, etc. If used for examination, it will affect the fairness and accuracy of the examination results. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a method and device for judging volleyball passing and padding ball training based on visual technology, which realizes real-time accurate counting based on computer vision technology combined with passing and padding ball state control, and improves the fairness and accuracy of the examination results.

[0005] In a first aspect, the present application provides a method for judging volleyball passing and padding ball training based on visual technology, comprising:

[0006] Deep learning model detection process: detecting target frame and finger key point in real time through a deep learning model, wherein the target frame includes a person frame, an arm frame, a hand frame and a ball frame, and the top of the person frame is the top point of the head;

[0007] Volleyball passing and padding ball judgment process:

[0008] According to the relationship between the finger key point coordinates and the person frame coordinates, it is judged whether the two hands are above the head;

[0009] When the two hands are above the head, the IOU of the hand frame and the ball frame is greater than a first set value, the hand touches the ball is detected, and then the ball flies upward, it is judged that the passing is successful;

[0010] When the two hands are not above the head, the IOU of the arm frame and the ball frame is greater than a first set value, the IOU of the two hand frames is greater than a second set value, and the ball flies above the head, it is judged that the padding is successful;

[0011] The process of logical judgment of passing and padding the ball is as follows:

[0012] After the start signal is sent, the image data of each frame is detected in real time, and the motion state flag is set to an initial value, and the number of successful passing and padding of the ball is set to 0;

[0013] After the successful padding of the ball is detected, the value of the motion state flag is read: if the initial value, the motion state flag is set to a first state value; if the second state value, the motion state flag is set to the first state value, the number of successful passing and padding of the ball is added by 1, and then the motion state flag is set to the initial value; if the first state value, the detection is continued;

[0014] After the successful passing of the ball is detected, the value of the motion state flag is read: if the initial value, the motion state flag is set to a second state value; if the first state value, the motion state flag is set to the second state value, the number of successful passing and padding of the ball is added by 1, and then the motion state flag is set to the initial value; if the second state value, the detection is continued;

[0015] If the ball falls or the timing time is up, the motion state flag is set to the initial value and the training is ended; when the training is ended, the detection is stopped, and the counting result of the number of successful passing and padding of the ball is output.

[0016] In a second aspect, the present application provides a device for judging the passing and padding of the ball in volleyball training based on visual technology, comprising:

[0017] a deep learning model detection module, configured to detect a target frame and finger key points in real time through a deep learning model, wherein the target frame comprises a person frame, an arm frame, a hand frame and a ball frame, and the top of the person frame is the top point of the head;

[0018] a volleyball passing and padding judgment module, configured to:

[0019] determine whether the two hands are above the head according to the relationship between the finger key point coordinates and the person frame coordinates;

[0020] when the two hands are above the head, it is determined that the hand touches the ball when the IOU of the hand frame and the ball frame is greater than a first set value, and then the ball flies upward, and it is determined that the passing of the ball is successful;

[0021] when the two hands are not above the head, it is determined that the padding of the ball is successful when the IOU of the arm frame and the ball frame is greater than the first set value, the IOU of the two hand frames is greater than a second set value, and the ball flies above the head;

[0022] a passing and padding logic judgment module, configured to:

[0023] After the start signal is sent, the image data of each frame is detected in real time and the motion state flag is set to an initial value, and the number of successful passes of the cushion ball is set to 0;

[0024] After the successful pass of the cushion ball is detected, the value of the motion state flag is read: if the initial value, the motion state flag is set to the first state value; if the second state value, the motion state flag is set to the first state value, the number of successful passes of the cushion ball is increased by 1, and then the motion state flag is set to the initial value; if the first state value, continue to detect;

[0025] After the successful pass of the cushion ball is detected, the value of the motion state flag is read: if the initial value, the motion state flag is set to the first state value; if the second state value, the motion state flag is set to the first state value, the number of successful passes of the cushion ball is increased by 1, and then the motion state flag is set to the initial value; if the first state value, continue to detect;

[0026] If the ball falls or the timing time is up, the motion state flag is set to the initial value and the training is ended; when the training is ended, the detection is stopped, and the counting result of the number of successful passes of the cushion ball is output.

[0027] The technical scheme provided in the embodiment of the application has at least the following technical effects:

[0028] Based on computer vision technology, including target box IOU algorithm, key point coordinate judgment, etc., and combined with the state control of the cushion ball, real-time accurate counting is realized, the fairness and accuracy of the examination result are improved, and the possibility of human error is reduced.

[0029] The above description is only a summary of the technical scheme of the application, in order to more clearly understand the technical means of the application, the application can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0030] The application will be further described below with reference to the accompanying drawings and in conjunction with the embodiments.

[0031] Figure 1 The flowchart of the method in the embodiment of the application is shown;

[0032] Figure 2 The structural schematic diagram of the device in the embodiment of the application is shown. DETAILED DESCRIPTION

[0033] The embodiment of the application provides a method and device for judging volleyball pass training based on visual technology, realizes real-time accurate counting based on computer vision technology combined with the state control of the cushion ball, and improves the fairness and accuracy of the examination result.

[0034] The technical solutions in the embodiments of the present application have the following general idea:

[0035] The present application is based on computer vision technology, and uses a deep convolutional neural network model to identify people, volleyball, arms, and finger key point positions. Then, according to business logic, whether it is a legal pass or a ball pad action is determined, and if there are two consecutive successful passes and pads, the number of successful passes and pads is increased by 1, and so on until the examination is completed. The overall process includes three processes: a deep learning model detection process, real-time frame detection of target boxes and finger key points; a volleyball pass and pad judgment process, which determines whether the pass and pad are successful by the relationship between the key point coordinates and the target box coordinates; and a pass and pad logic judgment process, which can accurately count the number of successful passes and pads by setting the motion state flag and the pass and pad judgment logic.

[0036] Before introducing the specific embodiments, the counting rules for passing and padding the ball are as follows:

[0037] After hearing the "start" signal, the examinee starts from the self-throwing ball, and continuously completes the self-passing and self-padding action on the head in the "pad-pass" or "pass-pad" exchange way. Each time a group of pass and pad actions are completed, one is counted. The score corresponding to the number of completed groups within 60 seconds is counted into the total score. During the test, the examinee can adjust by continuous passing or padding, and the time is not stopped when adjusting the ball, and the number is not counted until the "pad-pass" or "pass-pad" exchange action appears, and the counting continues. When padding the ball, the hands and arms must contact the ball at the same time, and the padding height must be that the whole ball exceeds the top of the head; when passing the ball, the hands must contact the ball above the forehead at the same time to complete the passing action. If the action is interrupted during the completion of the action, such as the ball falling to the ground, holding the ball with the hand, etc., it is considered as an examination.

[0038] Passing ball definition: the hands are in contact with the ball in the direction of the head and hit the ball upwards.

[0039] Padding ball definition: the hands and arms must contact the ball at the same time, and the padding height must be that the whole ball exceeds the top of the head.

[0040] Each examinee has two examination opportunities, and the best score is counted into the total score. If the examinee has no score in both examinations, he / she can take a supplementary test.

[0041] Site description:

[0042] One camera, the height of the camera is 2.5 meters, and the camera shoots from the side of the examinee.

[0043] Embodiment one

[0044] The embodiment provides a method for judging volleyball passing and padding training based on visual technology, which comprises the following steps: Figure 1As shown, the method can include the following steps:

[0045] S1, a deep learning model detection process: real-time frame detection of target boxes and finger key points through a deep learning model, wherein the target boxes include a person box, an arm box, a hand box, and a ball box, and the top of the person box is the top point of the head (the person box at least includes the head, and can also include the head and the entire body).

[0046] The target box can be represented as the coordinates (including the x coordinate and the y coordinate) of the upper left corner and the length-width information of the box; the finger key point is represented as the x coordinate and the y coordinate of the point; and whether the two hands are above the head is determined according to the relationship between the y coordinate of the finger key point and the y coordinate of the upper left corner of the person box.

[0047] The detection of the finger key point is specifically the detection of the top end, the bottom end, and the two joint points of each finger, a total of 4 points. The picture data set is input into a deep convolutional neural network model for training to improve the recognition accuracy.

[0048] S2, a volleyball passing and padding ball judgment process:

[0049] Whether the two hands are above the head is determined according to the relationship between the finger key point coordinates and the person box coordinates; specifically, according to the relationship between the finger key point coordinates and the upper left corner coordinates of the person box, if the y coordinate of the finger is above the y coordinate of the upper left corner of the person box, it is determined that the two hands are above the head.

[0050] When the two hands are above the head, the IOU (intersection over union) between the hand box and the ball box is greater than a first set value (such as 0.2), the hand touches the ball (IOU is the intersection part of the area of the hand box and the area of the ball box divided by the union part of the area of the hand box and the area of the ball box), and then the ball flies upward, and it is determined that the passing is successful.

[0051] When the two hands are not above the head, the IOU between the arm box and the ball box is greater than the first set value (such as 0.2, used to determine whether the ball is touched), and the IOU between the two hand boxes is greater than a second set value (such as 0.8, used to determine whether the two hands overlap), and when the ball flies above the head (the y coordinate of the ball is above the y coordinate of the upper left corner of the person box), it is determined that the padding is successful.

[0052] S3, a passing and padding ball logic judgment process:

[0053] After the start signal is sent, the image data of each frame is detected in real time, and the motion state flag is set to an initial value, and the number of successful passing and padding is set to 0;

[0054] After detecting the successful pad ball, the value of the motion state flag is read: if it is the initial value, the motion state flag is set to the first state value; if it is the second state value, the motion state flag is set to the first state value, the successful pad ball times are increased by 1, and then the motion state flag is set to the initial value; if it is the first state value, the detection is continued;

[0055] After detecting the successful pass, the value of the motion state flag is read: if it is the initial value, the motion state flag is set to the second state value; if it is the first state value, the motion state flag is set to the second state value, the successful pad ball times are increased by 1, and then the motion state flag is set to the initial value; if it is the second state value, the detection is continued;

[0056] If the ball falls or the timing time is up, the motion state flag is set to the initial value and the training is ended; the detection is stopped when the training is ended, and the counting result of the successful pad ball times is output.

[0057] The counting of the successful pad ball times is realized in the program as follows:

[0058] At the beginning, the motion state flag is set to the initial value, and the successful pad ball times are set to 0;

[0059] If the pad ball is successful, the motion state flag is set to the first state value (for example, 1);

[0060] If the pass is successful, the motion state flag is set to the second state value (for example, 2);

[0061] When the first state value and the second state value appear in turn (1 appears first and then 2 appears, or 2 appears first and then 1 appears), the successful pad ball times are increased by 1, and then the motion state flag is restored to the initial value 0.

[0062] The logic flow is clear, and the state change is clear, and there is no flow disorder and the like.

[0063] In a specific embodiment, the examination process is as follows:

[0064] Step 1. After the "start" signal is sent, real-time detection of image data of each frame is started, the motion state flag is set to the initial value of status 0, and the successful pad ball times flag count is set to the initial value of 0.

[0065] The status value indicates that 0 represents the initial value, 1 represents the successful pad ball (i.e. the first state value), and 2 represents the successful pass (i.e. the second state value).

[0066] Step 2. After the examinee throws the ball, no matter whether it is "pad ball" or "pass", as long as it is detected that one action is successful, it is considered that the examination starts, and the following will start with the first time being "pad ball" successful.

[0067] Step 3. After the first judgment of the "pad ball" action is successful, set the motion state flag status to 1.

[0068] Step 4. The flow continues, after the "pad ball" action is successful, the trajectory of the ball is first upward and then downward, and the judgment branch of the ball falling is as follows:

[0069] Branch one: if a successful "passing" action is detected after the ball falls, and the status flag bit value is 1, it means that the last time was a successful "pad ball" action, and this time was a successful "passing" action, both actions were successfully completed at the same time, then the success count flag count is incremented by 1, and the motion state flag status is set to 0.

[0070] Branch two: if a successful "pad ball" action is detected after the ball falls, then no processing is done, and the flow continues.

[0071] Branch three: if the ball falls to the ground after the ball falls, then the status flag bit value is set to the initial value 0, and the test is ended.

[0072] Step 5. If the ball does not stop, the flow repeats the processing of the above steps until the timing time is up, and the current test is ended. When the count result is output, save the video throughout the process, and output the action analysis and suggestions.

[0073] Through computer vision technology, the passing and padding of the ball by the athlete can be monitored and analyzed in real time, improving the training efficiency of the athlete, and providing detailed data analysis and feedback to help the athlete better adjust and improve their own technology, helping students understand their own technology and actions, and improving their own sports level.

[0074] Based on the same inventive concept, the present application also provides a device corresponding to the method in embodiment one, which is described in detail in embodiment two.

[0075] Embodiment two

[0076] In this embodiment, a device for judging volleyball passing and padding training based on visual technology is provided, as shown in Figure 2 , comprising:

[0077] A deep learning model detection module is configured to detect target frames and finger key points in real time through a deep learning model, wherein the target frames include a person frame, an arm frame, a hand frame, and a ball frame, and the top of the person frame is the top point of the head.

[0078] A volleyball passing and padding judgment module is configured to:

[0079] determine whether the two hands are above the head according to the relationship between the finger key point coordinates and the human frame coordinates;

[0080] When the two hands are above the head, it is determined that the hand touches the ball when the IOU between the hand frame and the ball frame is greater than a first set value, and the ball flies upward, and it is determined that the pass is successful.

[0081] When the two hands are not above the head, it is determined that the pass is successful when the IOU between the arm frame and the ball frame is greater than the first set value, the IOU between the two hand frames is greater than a second set value, and the ball flies upward.

[0082] The pass and pad ball logic determination module is used for:

[0083] After the start signal is sent, the image data of each frame is detected in real time, and the motion state flag is set to an initial value, and the number of successful passes and pads is set to 0.

[0084] After detecting that the pass and pad is successful, the value of the motion state flag is read: if it is the initial value, the motion state flag is set to a first state value; if it is the second state value, the motion state flag is set to the first state value, the number of successful passes and pads is increased by 1, and then the motion state flag is set to the initial value; if it is the first state value, continue to detect.

[0085] After detecting that the pass is successful, the value of the motion state flag is read: if it is the initial value, the motion state flag is set to a second state value; if it is the first state value, the motion state flag is set to the second state value, the number of successful passes and pads is increased by 1, and then the motion state flag is set to the initial value; if it is the second state value, continue to detect.

[0086] If the ball falls or the timing time is up, the motion state flag is set to the initial value and the training is ended; when the training is ended, the detection is stopped, and the counting result of the number of successful passes and pads is output.

[0087] Specifically, the target frame is represented as the coordinates of the upper left corner and the length and width information of the frame, and the finger key point is represented as the x and y coordinates of the point. Whether the two hands are above the head is determined according to the relationship between the y coordinates of the finger key point and the y coordinates of the upper left corner of the human frame.

[0088] Further, the camera for obtaining the image is arranged on the side of the user; when the counting result is output, the whole process video is saved, and the action analysis and suggestions are output.

[0089] Preferably, detecting the finger key point specifically refers to detecting the top end, the bottom end and the two joint points of each finger.

[0090] Specifically, the first set value is 0.2, and the second set value is 0.8.

[0091] Since the device introduced in the embodiment two of the present application is the device used for implementing the method of the embodiment one of the present application, the specific structure and deformation of the device can be understood by the person skilled in the art based on the method introduced in the embodiment one of the present application, and thus will not be described here again. The device used for the method of the embodiment one of the present application belongs to the scope of the present application.

[0092] The present application is based on computer vision technology, including target frame IOU algorithm, key point coordinate judgment, etc., and realizes real-time accurate counting by combining with the state control of the transmission pad, improves the fairness and accuracy of the test results, reduces the possibility of human error, reduces the work intensity of the judges, avoids the fatigue and distraction of the judges. The technical scheme of the present application has the following improvements: 1. improve the fairness and accuracy of the test results, reduce the possibility of human error; 2. through computer vision technology, the transmission pad action of the athlete can be monitored and analyzed in real time, and the training efficiency of the athlete is improved; 3. computer vision technology can provide detailed data analysis and feedback, so as to help the athlete better adjust and improve his own technology, help the student understand his own technology and action, and improve his own sports level.

[0093] Those skilled in the art will appreciate that embodiments of the present application can be provided as methods, apparatus, or computer program products. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage media, etc.) having computer-usable program code embodied in the medium.

[0094] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one or more flows and / or blocks. Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0095] These computer program instructions can also be stored in a computer-readable storage medium that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable storage medium produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0096] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flow Figure 1 one or more processes and / or blocks Figure 1 the steps of the function specified in the one or more blocks.

[0097] Although the present application has been described in terms of specific embodiments, it is apparent that persons skilled in the art will be able to modify these specific embodiments and the embodiments described herein, without departing from the spirit of the present application. It is therefore contemplated to cover by the present application any and all modifications, variations or equivalents that fall within the scope of the present application.

Claims

1. A method for judging volleyball passing and setting training based on visual technology, characterized in that, include: Deep learning model detection process: The target bounding box and finger key points are detected in real time using a deep learning model. The target bounding box includes the human bounding box, arm bounding box, hand bounding box and ball bounding box, and the top of the human bounding box is the vertex of the head. Volleyball passing and ball handling judgment process: Determine whether the hands are above the head based on the relationship between the coordinates of the key points of the fingers and the coordinates of the human frame; When both hands are above the head, if the IOU between the hand frame and the ball frame is greater than the first set value, it is determined that the hand has touched the ball. Then the ball flies upward, and the pass is considered successful. When both hands are not above the head, if the IOU between the arm frame and the ball frame is greater than the first set value, and the IOU between the two hand frames is greater than the second set value, and the ball flies upward above the head, then the ball is judged to be successfully passed. The logical judgment process of passing and setting the ball: After the start signal is issued, the image data of each frame is detected in real time and the motion status flag is set to the initial value. At the same time, the number of successful passes is set to 0. After a successful pass is detected, the value of the motion status flag is read: if it is the initial value, the motion status flag is set to the first status value; if it is the second status value, the motion status flag is set to the first status value, the number of successful passes is incremented by 1, and then the motion status flag is set to the initial value; if it is the first status value, the detection continues. After a successful pass is detected, the value of the motion status flag is read: if it is the initial value, the motion status flag is set to the second status value; if it is the first status value, the motion status flag is set to the second status value, the number of successful passes is incremented by 1, and then the motion status flag is set to the initial value; if it is the second status value, the detection continues. If the ball lands or the timer expires, the motion status flag is set to the initial value and the training ends; when the training ends, the detection stops and the count of the number of successful passes is output.

2. The method according to claim 1, characterized in that: The target box is represented by the coordinates of the top left corner and the length and width of the box. The finger keypoint is represented by the x and y coordinates of that point. The relationship between the y coordinate of the finger keypoint and the y coordinate of the top left corner of the person box is used to determine whether the hands are above the head.

3. The method according to claim 1 or 2, characterized in that: The key points of finger detection specifically involve detecting the tip, base, and two joints of each finger.

4. The method according to claim 1, characterized in that: The camera used to acquire images is positioned to the side of the user; when outputting counting results, the entire video is saved, and motion analysis and suggestions are also output.

5. The method according to claim 1, characterized in that: The first setting is 0.2, and the second setting is 0.

8.

6. A device for judging volleyball passing and setting training based on visual technology, characterized in that, include: The deep learning model detection module is used to detect target boxes and finger keypoints in real time using a deep learning model. The target boxes include human boxes, arm boxes, hand boxes, and ball boxes, and the top of the human box is the vertex of the head. The volleyball passing and ball-digging judgment module is used for: Determine whether the hands are above the head based on the relationship between the coordinates of the key points of the fingers and the coordinates of the human frame; When both hands are above the head, if the IOU between the hand frame and the ball frame is greater than the first set value, it is determined that the hand has touched the ball. Then the ball flies upward, and the pass is considered successful. When both hands are not above the head, if the IOU between the arm frame and the ball frame is greater than the first set value, and the IOU between the two hand frames is greater than the second set value, and the ball flies upward above the head, then the ball is judged to be successfully passed. The pass-and-place logic judgment module is used for: After the start signal is issued, the image data of each frame is detected in real time and the motion status flag is set to the initial value. At the same time, the number of successful passes is set to 0. After a successful pass is detected, the value of the motion status flag is read: if it is the initial value, the motion status flag is set to the first status value; if it is the second status value, the motion status flag is set to the first status value, the number of successful passes is incremented by 1, and then the motion status flag is set to the initial value; if it is the first status value, the detection continues. After a successful pass is detected, the value of the motion status flag is read: if it is the initial value, the motion status flag is set to the second status value; if it is the first status value, the motion status flag is set to the second status value, the number of successful passes is incremented by 1, and then the motion status flag is set to the initial value; if it is the second status value, the detection continues. If the ball lands or the timer expires, the motion status flag is set to the initial value and the training ends; when the training ends, the detection stops and the count of the number of successful passes is output.

7. The apparatus according to claim 6, characterized in that: The target box is represented by the coordinates of the top left corner and the length and width of the box. The finger keypoint is represented by the x and y coordinates of that point. The relationship between the y coordinate of the finger keypoint and the y coordinate of the top left corner of the person box is used to determine whether the hands are above the head.

8. The apparatus according to claim 6 or 7, characterized in that: The key points of finger detection specifically involve detecting the tip, base, and two joints of each finger.

9. The apparatus according to claim 6, characterized in that: The camera used to acquire images is positioned to the side of the user; when outputting counting results, the entire video is saved, and motion analysis and suggestions are also output.

10. The apparatus according to claim 6, characterized in that: The first setting is 0.2, and the second setting is 0.8.

Images