Wireless gamepad stability analysis system based on image analysis

Abstract

The invention relates to the technical field of wireless gamepad detection, in particular to a wireless gamepad stability analysis system based on image analysis. The system comprises a data acquisition unit, an image processing unit, an image analysis unit and a conduction judgment unit. According to the method, the images displayed on the displays are captured in the process of operating the gamepad, and the obtained response video is subjected to video frame segmentation, so that the response time of each display to the instruction of the gamepad is accurately obtained, and meanwhile, the segmented images are subjected to capture of game character actions, so that the game character actions are more accurately obtained. The action condition of the game character on each display under the instruction is obtained, and the standard action template established by the image processing unit is compared with the actual action of the game character on each display, so that the difference between the action of the game character on the display and the standard action template is obtained; therefore, the response speed of the wireless gamepad in use can be obtained more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless game controller detection, in particular to a wireless game controller stability analysis system based on image analysis. Background Art

[0002] Game controllers have been developed into wired and wireless game controllers. The transmission speed of wired game controllers is faster than that of wireless game controllers, while wireless game controllers are not limited to the control of transmission lines during use, and users can use them more casually. Each has its own advantages and disadvantages. When using a game controller to play a game, the user clicks the buttons on the game controller to make the game character on the display perform corresponding operations. If the operation buttons on the game controller are delayed after being clicked, the game character on the display will have delayed movements, which will affect the user's gaming experience. Especially for wireless game controllers, the speed of signal transmission determines the gaming experience. Therefore, after the game controller is produced, the performance of the game controller needs to be tested to ensure its normal use.

[0003] Currently, when testing wireless game controllers, staff are required to connect the game controllers wirelessly, and then determine the delay of the wireless controllers in operating game characters. However, manual detection methods can only perform rough detection, but not accurate detection. As competition continues to increase, the delay requirements for game controllers are also getting higher and higher. Therefore, a detection system is needed to replace manual detection, so as to accurately detect the effect of game controller delay, thereby improving the experience of later use of the game controller. Summary of the invention

[0004] The purpose of the present invention is to provide a wireless game controller stability analysis system based on image analysis to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned purpose, a wireless game controller stability analysis system based on image analysis is provided, comprising a data acquisition unit, an image processing unit, an image analysis unit and a conduction judgment unit; The data acquisition unit is used to connect the circuit on the wireless handle circuit board, and connect the wireless handle circuit board to the three displays respectively. After the wireless handle circuit board is connected, the wireless handle circuit board is input with instructions, so that the wireless handle circuit board responds to the instructions, and the images on the three displays are captured after the wireless handle circuit board responds with the instructions. The image processing unit establishes a standard action template according to the official action of the game, receives the pictures captured by the data acquisition unit on the three displays, divides the received pictures into video frames, and then captures the character actions on the divided video frames; The image analysis unit compares the motion postures of the three displays according to the character motion captured by the image processing unit, and determines the motion response according to each time point; The conduction judgment unit receives the reaction time of the character's action from the image analysis unit for comparison, judges the character's action posture, and then judges the wireless controller according to the reaction time comparison result and the action posture.

[0006] As a further improvement of the technical solution, the data acquisition unit includes a line connection control module, an instruction output module and an instruction recording module; The line connection control module is used to control the circuit connection device for electrically connecting to the wireless handle circuit board, and after the circuit connection of the wireless handle circuit is completed, the wireless handle circuit is connected to the three displays, wherein the connection modes of the wireless handle circuit and the three displays are: wireless, wired, and a combination of wired and wireless; The command output module is used to transmit commands to the wireless controller circuit board, so that the wireless controller circuit board triggers corresponding operations; The instruction recording module records the instructions output by the instruction output module, and also records the time point of the instruction output when recording.

[0007] As a further improvement of the technical solution, the data acquisition unit further includes a video motion capture module, and the image processing unit includes a standard motion template module; After the instruction output module outputs the instruction, the video motion capture module captures the motion of the image videos on the three displays. At the same time, the video motion capture module also captures the character motions provided by the game official, and after the video motion capture is completed, it is transmitted to the image processing unit; The standard action template module is used to define the video of the character action captured by the video action capture module as a standard action, and to make a template for the video content.

[0008] As a further improvement of the technical solution, the image processing unit further includes a video frame segmentation module and a motion state capture module; The video frame segmentation module is used to receive the image video displayed on the three displays captured by the video motion capture module, and after receiving, segment the image video according to the video frame so that the video forms a single picture; The action state capturing module recognizes and captures the actions of the characters in the pictures segmented by the video frame segmentation module.

[0009] As a further improvement of the technical solution, the step of the video frame segmentation module segmenting the image video into video frames is as follows: S1. First, the images and videos obtained on the three displays are numbered to distinguish the three images and videos; S2, reading a number of video frames in each image video, and numbering the images in each video frame according to the corresponding time; S3, according to the position of the video frame, the graphics displayed in each video frame are extracted separately, and a plurality of images are formed at this time; S4. Use the Canny edge detection operator algorithm to obtain the outline of the person in each image, and arrange the obtained outlines of several images according to the image numbers.

[0010] As a further improvement of the technical solution, the steps of the algorithm of the Canny edge detection operator are: ①. Gaussian filtering of the image: Use a one-dimensional Gaussian function to low-pass filter and remove noise from the original image f(x, y). The formula for Gaussian filtering is: , is the standard deviation of the Gaussian filter function, which controls the degree of smoothing. When is small, the filter positioning accuracy is high, but the signal-to-noise ratio is low; when When it is larger, the filter situation is just the opposite, and we need to determine the size according to the actual situation; ② Use the directional gradient operator to calculate the directional gradient of the image and the direction of the gradient: Set up two templates and , and perform convolution operation on the image through the above two templates to obtain the gradient values ​​of the pixels in the image in the x and y directions; Then set is the gradient in the x direction, is the gradient in the y direction. When calculating the amplitude, the L1 norm (sum of absolute values) is used, that is, ; Then calculate the gradient direction ; ③、Perform non-maximum suppression: The gradient direction of the edge is compared with different neighboring pixels in four directions: horizontal, vertical, 45° and 135°, to determine the local maximum. If the gray value of a pixel is not the largest compared with the gray values ​​of the two pixels before and after it in the gradient direction, the point is a non-edge point and is deleted; otherwise, it is a candidate edge point and is retained. This process is called "non-maximum suppression; ④. Use dual threshold technology for edge hysteresis: Use high and low thresholds to detect the results of non-maximum suppression to obtain high-value edge images and low-threshold edge images. Then, based on the images, use the images to supplement the edges of the connected images. ⑤. Get the edge of the image.

[0011] As a further improvement of the technical solution, the image analysis unit includes a frame duration comparison module, an action posture comparison module and an action reaction judgment module; The frame duration comparison module receives the segmented video images from the video frame segmentation module, obtains the video frames occupied by the image displayed on each display to complete the action, and calculates the time used to complete the specified action based on the number of video frames, and compares the time used to complete the specified action on each display; The action posture comparison module receives the action image captured by the action state capture module and the standard action template data in the standard action template module, and then compares the video frame of the completed action with the corresponding standard action template; The action reaction judgment module receives the results of the frame duration comparison module and the action posture comparison module on the video frame and the action posture, and then determines the posture and movement of the running action according to the compared data to judge the difference between the image and the standard action.

[0012] As a further improvement of the technical solution, the conduction judgment unit includes a reaction time comparison module, an action posture judgment module and a judgment module; The reaction time comparison module receives the time of completing the action on the corresponding display in the frame time comparison module and the time data of the instruction recorded in the instruction recording module, and then compares the time when the instruction is issued with the time corresponding to the video frame to determine the reaction time of starting the movement and completing the action on each display, and compares the reaction time between the three displays at the same time; The action posture judgment module receives the result of the action judgment by the action reaction judgment module, and after receiving, arranges the contours corresponding to the images of the image video segmentation on the three displays in the order of the video frames to form a video of the action contour, and then combines the video of the action contour with the video of the standard action, and then compares the action contour videos corresponding to the three displays with the standard action video for corresponding actions to determine the degree of distortion of the action posture; The evaluation module combines the comparison result of the reaction time by the reaction time comparison module and the comparison result of the action posture in the action posture judgment module to evaluate the operating performance of the wireless handle circuit board, and after an action evaluation is completed, it feeds back to the line connection control module, so that the line connection control module outputs the instruction for the next action according to the instruction of the instruction output module.

[0013] Compared with the prior art, the present invention has the following beneficial effects: In the wireless game controller stability analysis system based on image analysis, the game controller is connected by wire, wireless, or wired and wireless through a data acquisition unit, and commands are output to operate the game controller. In the process of operating the game controller, the image displayed on the display is captured to obtain the reaction videos of the game characters on three displays, and a standard action template is established through an image processing unit, and the reaction video obtained by the data acquisition unit is segmented into video frames, so as to accurately obtain the time for each display to react to the command of the game controller. At the same time, the image processing unit captures the action of the game character on the segmented image to obtain the action of the game character on each display under the command, and the standard action template established by the image processing unit is compared with the actual action of the game character on each display through the image analysis unit, so as to obtain the difference between the action of the game character on the display and the standard action template, and the transmission judgment unit is combined with the action reaction time and the action posture to judge the delayed transmission of the game controller, and the reaction of the wireless game controller is ensured according to the connection mode of wire, wireless, and wired and wireless, so as to more accurately obtain the reaction speed of the wireless game controller, and ensure that each wireless game controller is qualified before leaving the factory. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is the overall framework structure diagram of the present invention; Figure 2 A detailed structural diagram of the overall framework of the present invention; Figure 3 It is a demonstration block diagram of the present invention.

[0015] The meaning of each number in the figure is: 1. Data acquisition unit; 11. Line connection control module; 12. Command output module; 13. Video motion capture module; 14. Command recording module; 2. Image processing unit; 21. Standard action template module; 22. Video frame segmentation module; 23. Action state capture module; 3. Image analysis unit; 31. Frame duration comparison module; 32. Action posture comparison module; 33. Action reaction judgment module; 4. Conduction judgment unit; 41. Reaction time comparison module; 42. Action posture judgment module; 43. Evaluation module. DETAILED DESCRIPTION

[0016] 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.

[0017] Embodiment 1 See also Figure 1-Figure 3 As shown, a wireless game controller stability analysis system based on image analysis is provided, comprising a data acquisition unit 1, an image processing unit 2, an image analysis unit 3 and a conduction judgment unit 4; The data acquisition unit 1 is used to connect the circuit on the wireless handle circuit board, and connect the wireless handle circuit board to the three displays respectively. After the wireless handle circuit board is connected, the wireless handle circuit board is input with instructions, so that the wireless handle circuit board responds to the instructions, and the images on the three displays are captured after the wireless handle circuit board responds with the instructions. The data acquisition unit 1 includes a line connection control module 11, an instruction output module 12 and an instruction recording module 14; The line connection control module 11 is used to control the circuit connection device for electrically connecting to the wireless handle circuit board, and after the circuit connection of the wireless handle circuit is completed, the wireless handle circuit is connected to three displays, wherein the wireless handle circuit and the three displays are connected in the following ways: wireless, wired, and a combination of wired and wireless, wherein the wireless is the detection item, and the wired and the combination of wired and wireless are used for comparison with the wireless, the background displayed on the display is a solid color background, the game character is in the solid color background, and the color of the solid color background is a color that is easy for the computer to extract, such as white; After the wireless handle circuit board is placed on the circuit connection device, there are metal contacts on the circuit connection device that match the positions of the buttons on the wireless handle circuit board. When the circuit connection device and the wireless handle circuit board are electrically connected, the metal contacts are in contact with the wireless handle circuit board. The command output module 12 is used to transmit commands to the wireless controller circuit board, so that the wireless controller circuit board triggers corresponding operations, that is, the command output module 12 sends electrical signals to the wireless controller circuit board through metal contacts, and the sent electrical signals are used to control the movement of the game character on the display; The instruction recording module 14 records the instruction output by the instruction output module 12, and also records the time point of the instruction output when recording. The recorded time is the initial time of this detection, that is, 0 hours. After the instruction output module 12 sends the instruction signal, the time recording starts; The data acquisition unit 1 also includes a video motion capture module 13. After the instruction output module 12 outputs the instruction, the video motion capture module 13 captures the motion of the image videos on the three displays. The capture method uses a camera to capture, which is simple to operate and has low cost. At the same time, the video motion capture module 13 also captures the character motion provided by the game official, and after the video motion capture is completed, it is transmitted to the image processing unit 2; The image processing unit 2 establishes a standard action template according to the official action of the game, receives the images captured by the data acquisition unit 1 on the three displays, and divides the received images into video frames, and then captures the character actions on the divided video frames; The image processing unit 2 includes a standard action template module 21; The standard action template module 21 is used to define the video of the character action captured by the video action capture module 13 as a standard action, and to make a template for the video content. By establishing a template for the character action, the character action video captured by the video action capture module 13 has a reference record, so as to determine whether the character action displayed on the display is standard.

[0018] The image processing unit 2 also includes a video frame segmentation module 22 and a motion state capture module 23; The video frame segmentation module 22 is used to receive the image video displayed on the three displays captured by the video motion capture module 13, and after receiving the image video, segment the image video according to the video frame so that the video forms a single picture; The steps of the video frame segmentation module 22 segmenting the image video into video frames are as follows: S1. First, number the images and videos obtained on the three displays to distinguish the three images and videos, such as numbering the images and videos obtained on the three displays as A, B, and C respectively; S2, read a number of video frames in each image video, and at the same time number the images in each video frame corresponding to the time. Taking the image video numbered A as an example, when a number of video frames are read out from the image video A and numbered, such as the image of the nth frame and the time m when the instruction is issued by the instruction output module 12, the number is Anm; S3, according to the position of the video frame, the graphics displayed in each video frame are extracted separately, and a plurality of images are formed at this time; S4, using the Canny edge detection operator algorithm to obtain the outline of the person in each image, and arranging the obtained outlines of several images according to the image numbers; The steps of the Canny edge detection operator algorithm are: ①. Gaussian filtering of the image: Use a one-dimensional Gaussian function to low-pass filter and remove noise from the original image f(x, y). The formula for Gaussian filtering is: , is the standard deviation of the Gaussian filter function, which controls the degree of smoothing. When is small, the filter positioning accuracy is high, but the signal-to-noise ratio is low; when When it is larger, the filter situation is just the opposite, and we need to determine the size according to the actual situation; ② Use the directional gradient operator to calculate the directional gradient of the image and the direction of the gradient: Set up two templates and , and perform convolution operation on the image through the above two templates to obtain the gradient values ​​of the pixels in the image in the x and y directions; Then set is the gradient in the x direction, is the gradient in the y direction. When calculating the amplitude, the L1 norm (sum of absolute values) is used, that is, ; Then calculate the gradient direction ; ③、Perform non-maximum suppression: The gradient direction of the edge is compared with different neighboring pixels in four directions: horizontal, vertical, 45° and 135°, and the local maximum is determined. If the gray value of a pixel is not the largest compared with the gray values ​​of the two pixels before and after it in the gradient direction, the point is a non-edge point and is deleted; otherwise, it is a candidate edge point and is retained; ④. Use dual threshold technology for edge hysteresis: Use high and low thresholds to detect the results of non-maximum suppression to obtain high-value edge images and low-threshold edge images. Then, based on the image, use the image to supplement the edge of the image. After the edge lines on the image are supplemented, the edge of the image will gradually become complete. When the edge of the image is supplemented, the image processing is completed. ⑤. Get the edge of the image: The posture of the game character in the image is obtained by obtaining the edge of the image, and the action of the character in the frame image is obtained by the posture of the game character; The action state capturing module 23 recognizes and captures the human action on the image segmented by the video frame segmentation module 22; The process of capturing the action of the character by the action state capturing module 23 is to identify the edge of the image obtained by the video frame segmentation module 22, and after identifying the edge of the image, the edge lines of the image are captured and extracted, and after extraction, the edge lines of the image are numbered, and the numbering is continued as Anm-line based on the above-mentioned Anm; The image analysis unit 3 compares the motion postures of the three displays according to the character motion captured by the image processing unit 2, and determines the motion response according to each time point; The image analysis unit 3 includes a frame duration comparison module 31, an action posture comparison module 32 and an action reaction judgment module 33; The frame duration comparison module 31 receives the segmented video images received by the video frame segmentation module 22, obtains the video frames occupied by the image displayed on each display to complete the action, and calculates the time used to complete the specified action according to the number of video frames, and compares the time used by each display to complete the specified action, that is, the time recorded on the instruction recording module 14 when a certain action in video A appears, the time recorded on the instruction recording module 14 when a certain action in video B appears, and the time difference between the time recorded on the instruction recording module 14 when a certain action in video C appears, and obtains the speed of the instruction transmission of the three methods of connecting to the wireless handle circuit board by comparing the time; The action posture comparison module 32 receives the action image captured by the action state capture module 23 and the standard action template data in the standard action template module 21, and then compares the video frame of the completed action with the corresponding standard action template; Here, the process of comparing the posture of the video frame and the corresponding standard action template is as follows: The lines of the image captured by the motion state capturing module 23 are transmitted to the motion posture comparing module 32. At this time, the lines of the image transmitted to the motion posture comparing module 32 by the motion state capturing module 23 are all the lines of the video images obtained on the three displays; Then, the standard action template of the standard action template module 21 is transmitted to the action posture comparison module 32, and the image lines of the characters starting to perform actions obtained on the three displays and the standard action template are placed together, and the lines corresponding to each video frame are matched. The matching points here are not related to time. For example, when the game handle controls the game character to swing his arms, when the action posture comparison module 32 performs action posture comparison, it is any action in the arm swinging work, and they are all corresponding, that is, the actions corresponding to the three displays are the initial actions of raising the arms first, and the standard action template is also this action. By comparing the four corresponding actions together, the action posture comparison module 32 can quickly compare the differences between the actions of the characters displayed on the three displays and the actions of the standard action template; The action reaction judgment module 33 receives the results of the video frame and action posture comparison by the frame duration comparison module 31 and the action posture comparison module 32, and then determines the posture and motion of the running action based on the compared data to judge the difference between the image and the standard action. When the action reaction judgment module 33 makes a judgment, it compares the pictures of each action.

[0019] The transmission judgment unit 4 receives the reaction time of the character's action from the image analysis unit 3, compares it, and judges the character's action posture, and then judges the wireless controller according to the reaction time comparison result and the action posture; The conduction judgment unit 4 includes a reaction time comparison module 41, an action posture judgment module 42 and a judgment module 43; The reaction time comparison module 41 receives the time of completing the action on the corresponding display in the frame time comparison module 31 and the time data of the instruction recorded in the instruction recording module 14, and then compares the time when the instruction is issued with the time corresponding to the video frame to determine the reaction time of starting the movement and completing the action on each display, and compares the reaction time between the three displays at the same time. In this way, the total time for each display to complete the action corresponding to the instruction is first obtained, and the time consumed by the corresponding character in the display to perform each action can also be obtained. The consumed time is the time difference between the time when the action is displayed on the display and the time when the instruction is issued; The action posture judgment module 42 receives the result of the action judgment by the action reaction judgment module 33, and after receiving, arranges the contours corresponding to the images of the image video segmentation on the three displays in the order of the video frames to form a video of the action contour, and then combines the video of the action contour with the video of the standard action, and then compares the action contour videos corresponding to the three displays with the standard action video for corresponding actions to determine the degree of distortion of the action posture; The judging module 43 judges the operating performance of the wireless handle circuit board by combining the comparison result of the reaction time by the reaction time comparison module 41 and the comparison result of the action posture in the action posture judgment module 42. At this time, the judging module 43 judges the difference between the wired and wireless by combining the data in the reaction time comparison module 41 and the action posture judgment module 42 to make a judgment on the difference between the wired and wireless. When the judged difference is within the qualified range specified by the company, it means that the electronic components installed on the wireless handle circuit board are qualified and meet the factory standards of the wireless handle circuit board. After an action judgment is completed, it is fed back to the line connection control module 11, so that the line connection control module 11 outputs the command for the next action according to the command of the command output module 12. By repeating the test several times, the test effect of the wireless handle circuit board is improved to ensure that each wireless handle circuit board shipped from the factory is qualified.

[0020] The data acquisition unit 1 is used to connect the game controller by wire, wireless, or wired and wireless, and to output instructions to operate the game controller. In the process of operating the game controller, the image displayed on the display is captured to obtain the reaction video of the game character on the three displays. The image processing unit 2 is used to establish a standard action template, and the reaction video obtained by the data acquisition unit 1 is segmented into video frames, so as to accurately obtain the time for each display to respond to the instruction of the game controller. At the same time, the image processing unit 2 captures the action of the game character on the segmented image to obtain the action of the game character on each display under the instruction. The image analysis unit 3 compares the standard action template established by the image processing unit 2 with the actual action of the game character on each display to obtain the difference between the action of the game character on the display and the standard action template. The conduction judgment unit 4 is used to judge the delayed transmission of the game controller in combination with the action reaction time and the action posture. The reaction of the wireless game controller is ensured by connecting in a wired, wireless, or wired and wireless manner, so as to more accurately obtain the reaction speed of the wireless game controller, and ensure that each wireless game controller is qualified before leaving the factory.

[0021] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A wireless game controller stability analysis system based on image analysis, characterized by: It comprises a data acquisition unit (1), an image processing unit (2), an image analysis unit (3) and a conduction judgment unit (4); The data acquisition unit (1) is used to connect the circuit on the wireless handle circuit board and connect the wireless handle circuit board to the three displays respectively. After the wireless handle circuit board is connected, a command is input to the wireless handle circuit board so that the wireless handle circuit board responds to the command accordingly. After the wireless handle circuit board responds to the command, the images on the three displays are captured. The image processing unit (2) establishes a standard action template according to the official action of the game, receives the images captured by the data acquisition unit (1) on the three displays, divides the received images into video frames, and then captures the character actions in the divided video frames; The image analysis unit (3) compares the action postures of the three displays according to the character actions captured by the image processing unit (2), and judges the action response according to each time point; The conduction judgment unit (4) receives the image analysis unit (3) to compare the reaction time of the character's action, judges the character's action posture, and then judges the wireless controller based on the reaction time comparison result and the action posture.

2. The wireless game controller stability analysis system based on image analysis according to claim 1, characterized in that: The data acquisition unit (1) comprises a line connection control module (11), an instruction output module (12) and an instruction recording module (14); The line connection control module (11) is used to control a circuit connection device for electrically connecting to a wireless handle circuit board, and after the circuit connection of the wireless handle circuit is completed, connect the wireless handle circuit to three displays, wherein the wireless handle circuit and the three displays are connected in the following ways: wireless, wired, and a combination of wired and wireless; The command output module (12) is used to transmit commands to the wireless handle circuit board, so that the wireless handle circuit board initiates corresponding operations; The instruction recording module (14) records the instructions output by the instruction output module (12), and also records the time point when the instructions are output.

3. The wireless game controller stability analysis system based on image analysis according to claim 2, characterized in that: The data acquisition unit (1) further comprises a video motion capture module (13), and the image processing unit (2) comprises a standard motion template module (21); After the instruction output module (12) outputs the instruction, the video motion capture module (13) captures the motion of the image videos on the three displays. At the same time, the video motion capture module (13) also captures the character motion provided by the game official, and after the video motion capture is completed, it is transmitted to the image processing unit (2); The standard action template module (21) is used to define the video of the character action captured by the video action capture module (13) as a standard action, and to create a template based on the video content.

4. The wireless game controller stability analysis system based on image analysis according to claim 3 is characterized in that: The image processing unit (2) further comprises a video frame segmentation module (22) and a motion state capture module (23); The video frame segmentation module (22) is used to receive the image video displayed on three displays captured by the video motion capture module (13), and after receiving the image video, segment the image video according to the video frame so that the video forms a single picture; The action state capturing module (23) recognizes and captures the actions of the characters in the pictures segmented by the video frame segmentation module (22).

5. The wireless game controller stability analysis system based on image analysis according to claim 4, characterized in that: The steps of segmenting the video frames of the image video by the video frame segmentation module (22) are as follows: S1. First, the images and videos obtained on the three displays are numbered to distinguish the three images and videos; S2, reading a number of video frames in each image video, and numbering the images in each video frame according to the corresponding time; S3, according to the position of the video frame, the graphics displayed in each video frame are extracted separately, and a plurality of images are formed at this time; S4. Use the Canny edge detection operator algorithm to obtain the outline of the person in each image, and arrange the obtained outlines of several images according to the image numbers.

6. The wireless game controller stability analysis system based on image analysis according to claim 5, characterized in that: The steps of the algorithm of the Canny edge detection operator are: ①, perform Gaussian filtering on the image; ② Use the directional gradient operator to calculate the directional gradient of the image and the direction of the gradient; ③. Perform non-maximum suppression; ④. Use dual threshold technology for edge hysteresis; ⑤. Get the edge of the image.

7. The wireless game controller stability analysis system based on image analysis according to claim 4, characterized in that: The image analysis unit (3) comprises a frame duration comparison module (31), an action posture comparison module (32) and an action reaction judgment module (33); The frame duration comparison module (31) receives the segmented video images from the video frame segmentation module (22), obtains the video frames occupied by the image displayed on each display to complete the action, calculates the time taken to complete the specified action based on the number of video frames, and compares the time taken by each display to complete the specified action; The action posture comparison module (32) receives the action image captured by the action state capture module (23) and the standard action template data in the standard action template module (21), and then compares the posture of the video frame of the completed action with the corresponding standard action template; The action reaction judgment module (33) receives the results of the comparison of the video frame and the action posture by the frame duration comparison module (31) and the action posture comparison module (32), and then determines the posture and movement of the running action based on the compared data to judge the difference between the image and the standard action.

8. The wireless game controller stability analysis system based on image analysis according to claim 7, characterized in that: The conduction judgment unit (4) comprises a reaction time comparison module (41), an action posture judgment module (42) and a judgment module (43); The reaction time comparison module (41) receives the time of completing the action on the corresponding display in the frame time comparison module (31) and the time data of the instruction recorded in the instruction recording module (14), and then compares the time when the instruction is issued with the time corresponding to the video frame to determine the reaction time of starting the movement and completing the action on each display, and compares the reaction time between the three displays at the same time; The action posture judgment module (42) receives the result of the action judgment by the action reaction judgment module (33), and after receiving the result, arranges the contours corresponding to the images of the image video segmentation on the three displays in the order of the video frames to form a video of the action contour, and then combines the video of the action contour with the video of the standard action, and then compares the action contour videos corresponding to the three displays with the standard action video for the corresponding actions to determine the degree of distortion of the action posture; The evaluation module (43) evaluates the operating performance of the wireless handle circuit board by combining the comparison result of the reaction time by the reaction time comparison module (41) and the comparison result of the action posture by the action posture judgment module (42), and after the evaluation of an action is completed, feedback is given to the line connection control module (11), so that the line connection control module (11) performs the instruction output of the next action according to the instruction of the instruction output module (12).

Images