A method for automatic counting of skipping rope based on artificial intelligence

Through artificial intelligence, the video data is analyzed and the number of skipping ropes is automatically counted, which solves the errors and data unfair problems that are easily caused by traditional manual counting, and realizes the accuracy and fairness of skipping rope counting.

CN116563922BActive Publication Date: 2024-11-01NANJING LINGYUE QIANCHENG INTELLIGENT TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310533219.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-12
Publication Date
2024-11-01
Estimated Expiration
2043-05-12

AI Technical Summary

Technical Problem

The traditional rope skipping counting method requires manual assistance, which is prone to errors due to fatigue of counting personnel, and the statistics of rope skipping movements with smaller amplitude are unfair.

Method used

Using an artificial intelligence-based method, the human body posture information is captured through video data acquisition equipment, the Cartesian coordinate system is used to record the position information of the ankle and shoulders, and the jumping action is analyzed and judged in real time, and the number of jumping ropes is automatically counted, and the maximum value of the y2 coordinates of the shoulders is passed and Comparison of the average value determines the jump rope amplitude to ensure accurate counting.

Benefits of technology

The error problem caused by manual counting is solved, the fairness and accuracy of the rope skipping data is ensured, and the errors caused by artificial fatigue are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116563922B_ABST
    Figure CN116563922B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field related to intelligent fitness exercise, and more specifically, the present invention provides an automatic rope skipping counting method based on artificial intelligence, comprising the following steps: S1: real-time video information is collected by a video data acquisition device, and the video information is displayed by a screen of a display device; S2: faces in the video are captured, and after the person to be tested is determined according to the captured face information, key points of the human posture information of the test person are identified, and the key points of the human posture information include the ankles and shoulders of the human body; and the position information of the key points of the human posture information of the test person is collected in real time; it is possible to determine whether the test person is in a test area by fixing the ankle position, and then the rope skipping is automatically counted by collecting the change of the y2 coordinate of the test person's shoulder during the rope skipping process, thereby solving the problem that counting errors are easily caused by fatigue of the counting personnel in the method of manually counting rope skipping.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field related to intelligent fitness exercise, and more specifically, to an automatic rope skipping counting method based on artificial intelligence. Background Art

[0002] With the continuous maturity and development of artificial intelligence technology and big data analysis technology, it has become a trend for artificial intelligence to replace traditional physical education. Traditional rope skipping counting uses manual counting to enter scores, which requires manual assistance technology and takes a lot of time to sort out the data. In addition, when the number of rope skipping is large, it is easy to cause counting errors due to fatigue of the counting personnel. In addition, the statistics of rope skipping with smaller amplitudes are counted in the number of rope skipping, which will lead to unfair statistical data. Summary of the invention

[0003] The purpose of the present invention is to provide an automatic rope skipping counting method based on artificial intelligence to solve the problems raised in the above background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for automatically counting rope skipping based on artificial intelligence, comprising the following steps:

[0005] S1: Real-time video information is collected through a video data acquisition device, and the video information is displayed on a screen of a display device;

[0006] S2: Capture the face in the video, determine the person to be tested based on the captured face information, and then identify the key points of the test person's human posture information, which include the ankles and shoulders of the human body; collect the position information of the key points of the test person's human posture information in real time, and analyze the human posture information based on the position information of the key points of the human posture information to determine whether the test person is located in the test area. When the key points of the test person's human body are all located in the test area, it is determined that the test person is located in the test area; the method for real-time collection of the position information of the key points of the human posture information includes taking a corner in the test area as the origin, the horizontal direction as the x-axis, and the vertical direction as the y-axis to form a Cartesian coordinate system, recording the real-time (x, y1) coordinates of the ankle to complete the collection of the ankle position information; recording the real-time y2 coordinate of the shoulder to complete the collection of the shoulder position information;

[0007] S3: During the test time, multiple different y2 coordinates of the shoulders of the tester in a single rope skipping in the test area are collected, and the maximum value of the y2 coordinate of the shoulders collected by the tester in a single rope skipping in the test area is extracted. Based on the y3 coordinate of the average position, the maximum value of the y2 coordinate of the shoulders in multiple rope skipping is compared with the average value y3 coordinate of the rope skipping respectively. When the maximum value of the y2 coordinate of the shoulders in the rope skipping action exceeds 45%-55% of the y3 coordinate of the average position, the rope skipping count is increased by 1;

[0008] S4: After the tester confirms that the rope skipping is completed, the number of rope skipping is output.

[0009] A further technical solution of the present application: In step S3, the method for determining the average position includes:

[0010] S31: forming a set of calculation intervals according to multiple y2 coordinates of the shoulder position collected by the tester in a single rope skipping, and calculating the median of the calculation interval according to the maximum value and the minimum value in the calculation interval;

[0011] S32: Calculate the average value based on the median value obtained by the tester's multiple rope skipping, and the obtained average value is the y coordinate of the average position.

[0012] A further technical solution of the present application: the automatic rope skipping counting method based on artificial intelligence further comprises the following steps:

[0013] S5: The duration, number of jumps and y`3 coordinate of the historical average position of each skipping test are stored in the personal file of the corresponding tester, and the skipping speed of the tester is analyzed according to the duration and number of jumps of a single skipping test.

[0014] A further technical solution of the present application: In step S3, the method for determining the average position includes:

[0015] S31: forming a set of calculation intervals according to multiple y2 coordinates of the shoulder position collected by the tester in a single rope skipping, and calculating the median of the calculation interval according to the maximum value and the minimum value in the calculation interval;

[0016] S32: Calculate the average value based on the median value obtained by the tester's multiple rope skipping and the y'3 coordinate of the historical average position stored in the tester's file, and the obtained average value is the y3 coordinate of the current average position.

[0017] A further technical solution of the present application is as follows: in step S2, after determining the person to be tested based on the captured facial information, the video interface displayed on the display device screen is cut so that only the key points of all the body posture information of the test person appear in the video interface.

[0018] A further technical solution of the present application: in step S2, when it is determined that the test person is located in the test area, the rope skipping test start information is output.

[0019] A further technical solution of the present application is as follows: in step S2, when it is determined that the test person is located in the test area, the countdown information for the start of the rope skipping test is inserted in the form of a pop-up window in the screen of the display device. When the countdown to the start of the rope skipping test ends, the pop-up window is closed and the rope skipping actions of the test person begin to be counted.

[0020] A further technical solution of the present application is as follows: in step S2, when it is determined that the test person is located in the test area, after the voice information of the start of the rope skipping test is output through the voice output device, the rope skipping actions of the test person are started to be counted.

[0021] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0022] 1. The automatic rope skipping counting method based on artificial intelligence of the present invention can determine whether the tester is in the test area by fixing the ankle position, and then automatically count the rope skipping by collecting the change of the y2 coordinate of the tester's shoulder during the rope skipping process, thereby solving the problem of counting errors caused by fatigue of the counting personnel in the manual rope skipping counting method.

[0023] 2. The present invention uses the maximum value of the shoulder y2 coordinate collected in a single rope skipping and the y3 coordinate of the average position as a benchmark, and compares the maximum value of the shoulder y2 coordinate in multiple rope skipping with the average value y3 coordinate of the rope skipping. When the maximum value of the shoulder y2 coordinate in the rope skipping action exceeds 45%-55% of the average position y3 coordinate, it is determined that the amplitude of the rope skipping meets the standard, and the rope skipping count is increased by 1. This solves the problem of unfair statistical data because rope skipping actions with smaller amplitudes are counted in the number of rope skipping actions. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of an automatic rope skipping counting method based on artificial intelligence provided by the present invention;

[0025] Figure 2 A schematic diagram of collecting the position information of key points of human posture information provided by the present invention;

[0026] Figure 3 A schematic diagram of an interface for displaying image information of a test person skipping rope on a display device provided by the present invention;

[0027] Figure 4A schematic diagram of y2 coordinate data for identifying shoulder positions during multiple rope skipping within the test time provided by the present invention;

[0028] Figure 5 This is a schematic diagram of the amplitude information of the y2 coordinate of the shoulder position during multiple rope skipping processes within the test time provided by the present invention.

[0029] Explanation of the symbols in the schematic diagram: DETAILED DESCRIPTION

[0030] 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. The present invention is further described below in combination with the embodiments.

[0031] See also Figure 1-Figure 5 In one embodiment of the present application, a method for automatically counting rope skipping based on artificial intelligence comprises the following steps:

[0032] S1: Real-time video information is collected through a video data acquisition device, and the video information is displayed on a screen of a display device;

[0033] S2: Capture the face in the video, determine the person to be tested based on the captured face information, and then identify the key points of the test person's human posture information, which include the ankles and shoulders of the human body; collect the position information of the key points of the test person's human posture information in real time, and analyze the human posture information based on the position information of the key points of the human posture information to determine whether the test person is located in the test area. When the key points of the test person's human body are all located in the test area, it is determined that the test person is located in the test area; the method for real-time collection of the position information of the key points of the human posture information includes taking a corner in the test area as the origin, the horizontal direction as the x-axis, and the vertical direction as the y-axis to form a Cartesian coordinate system, recording the real-time (x, y1) coordinates of the ankle to complete the collection of the ankle position information; recording the real-time y2 coordinate of the shoulder to complete the collection of the shoulder position information;

[0034] S3: During the test time, multiple different y2 coordinates of the shoulders of the tester in a single rope skipping in the test area are collected, and the maximum value of the y2 coordinate of the shoulders collected by the tester in a single rope skipping in the test area is extracted. Based on the y3 coordinate of the average position, the maximum value of the y2 coordinate of the shoulders in multiple rope skipping is compared with the average value y3 coordinate of the rope skipping respectively. When the maximum value of the y2 coordinate of the shoulders in the rope skipping action exceeds 45% to 55% of the y3 coordinate of the average position, the rope skipping count is increased by 1;

[0035] S4: After the tester confirms that the rope skipping is completed, the number of rope skipping is output.

[0036] In one case of this embodiment, the video data acquisition device uses a 2 million zoom network camera that captures video images in real time, uses a router to provide data transmission, reads network camera data in real time through the RTSP protocol, detects human posture through the Tensorflow Lite posture recognition model in the prior art, locates the positions of ankles and shoulders, analyzes the posture information of the human body in the image, and tracks the fixed points of ankles and shoulders in real time. Thus, the skipping rope can be counted according to the y2 coordinate of the shoulder position.

[0037] See also Figure 4 and Figure 5 As a preferred embodiment of the present application, in step S3, the method for determining the average position includes:

[0038] S31: A set of calculation intervals {y 21 ,y 22 ,y 23 ,y 24 ,y 25 ……y 2i}, according to the maximum value y in the calculation interval 2i and the minimum value y 21 , calculate the median of the calculation interval

[0039] S32: n medians calculated based on the tester's n times of skipping To calculate the average value, the average value is the y coordinate of the average position

[0040] In order to further determine the y3 coordinate of the average value of the test persons, as a preferred embodiment of the present application, the automatic rope skipping counting method based on artificial intelligence further includes the following steps:

[0041] S5: The duration, number of jumps and y3 coordinate of the average position in each rope skipping test are stored in the personal file of the corresponding tester, and the speed of the tester's rope skipping is analyzed according to the duration and number of jumps of a single rope skipping.

[0042] In one case of this embodiment, in step S3, the method for determining the average position includes:

[0043] S31: A set of calculation intervals {y 21 ,y 22 ,y 23 ,y 24 ,y 25 ……y 2n}, according to the maximum value y in the calculation interval 2n and the minimum value y 21 , calculate the median of the calculation interval

[0044] S32: n medians calculated based on the tester's n times of skipping and the y`3 coordinates of the j historical average positions stored in the tester's profile {y` 31 ,y` 32 ,y` 33 ...y` 3j} to calculate the average value, and the average value obtained is the y3 coordinate of the current average position

[0045] See also Figure 2 and Figure 3 As a preferred embodiment of the present application, in step S2, after determining the person to be tested based on the captured facial information, the video interface displayed on the display device screen is cut so that only the key points of all the human posture information of the test person appear in the video interface.

[0046] In actual application, pixel points are obtained through the opencv function Mat (Mat m, Rect roi), so that the pixel points at different positions formed a rectangle, and the image data of the rectangular area can be obtained. It is necessary to explain that cutting an image through the opencv function is a prior art, and this embodiment will not be repeated here.

[0047] In order to further improve the preparations before rope skipping, as a preferred embodiment of the present application, in step S2, when it is determined that the test person is located in the test area, the rope skipping test start information is output.

[0048] In one case of the present embodiment, in step S2, when it is determined that the tester is located in the test area, the countdown information for the start of the rope skipping test is inserted in the form of a pop-up window in the screen of the display device. When the countdown for the rope skipping test ends, the pop-up window is automatically closed or the tester clicks on the close pop-up window of the display device and the counting of the tester's rope skipping actions begins.

[0049] In another case of this embodiment, in step S2, when it is determined that the test person is located in the test area, after the voice information of the start of the rope skipping test is output through the voice output device, the rope skipping actions of the test person are started to be counted.

[0050] In actual application, the voice output device can be an outdoor sound column, a speaker or a loudspeaker.

[0051] The steps of the present invention are as follows:

[0052] S1: Real-time video information is collected through a video data acquisition device, and the video information is displayed on a screen of a display device;

[0053] S2: Capture the face in the video, determine the person to be tested based on the captured face information, and then identify the key points of the test person's human posture information, which include the ankles and shoulders of the human body; collect the position information of the key points of the test person's human posture information in real time, and analyze the human posture information based on the position information of the key points of the human posture information to determine whether the test person is located in the test area. When the key points of the test person's human body are all located in the test area, it is determined that the test person is located in the test area; the method for real-time collection of the position information of the key points of the human posture information includes taking a corner in the test area as the origin, the horizontal direction as the x-axis, and the vertical direction as the y-axis to form a Cartesian coordinate system, recording the real-time (x, y1) coordinates of the ankle to complete the collection of the ankle position information; recording the real-time y2 coordinate of the shoulder to complete the collection of the shoulder position information;

[0054] S3: During the test time, multiple different y2 coordinates of the tester's shoulders during a single rope skipping in the test area are collected, and a set of calculation intervals {y 21 ,y 22 ,y 23 ,y 24 ,y 25 ……y 2i}, according to the maximum value y in the calculation interval 2i and the minimum value y 21 , calculate the median of the calculation interval Extract the maximum value y2i and minimum value y2i of the shoulder y2 coordinate collected by the tester in a single rope skipping in the test area 21 , calculate the median of the calculation interval The n median values ​​calculated based on the tester's n rope skipping The n median values ​​calculated based on the tester's n rope skipping and the y`3 coordinates of the j historical average positions stored in the tester's profile {y` 31 ,y` 32,y` 33 ……y 3j} to calculate the average value, and the average value obtained is the y3 coordinate of the current average position Taking the y3 coordinate of the average position as the benchmark, the maximum value of the y2 coordinate of the shoulder in multiple skipping ropes is compared with the average value y3 coordinate of the skipping rope. When the maximum value of the y2 coordinate of the shoulder in the skipping rope action exceeds 50% of the y3 coordinate of the average position, the amplitude of the skipping rope is judged to meet the standard, and the skipping rope count is increased by 1. If the maximum value of the y2 coordinate of the shoulder in the skipping rope action is within 50% of the y3 coordinate of the average position, the amplitude of the skipping rope is judged to not meet the standard, and the skipping rope count is neither increased nor decreased.

[0055] S4: After the tester confirms that the rope skipping is completed, the number of rope skipping is output.

[0056] S5: The duration, number of jumps and y'3 coordinates of the historical average position of each jump rope test are stored in the personal file of the corresponding tester, and the tester's jump rope speed is analyzed according to the duration and number of jumps of a single jump rope.

[0057] The present invention and its embodiments are described schematically above, and the description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by it and designs a structural method and an embodiment similar to the technical solution without creativity without departing from the purpose of the invention, they shall all fall within the protection scope of the present invention.

[0058] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for automatically counting rope skipping based on artificial intelligence, characterized in that: The following steps are involved: S1: Real-time video information is collected through a video data acquisition device, and the video information is displayed on a screen of a display device; S2: Capture the face in the video, determine the person to be tested based on the captured face information, identify the key points of the test person's human posture information, and the key points of the human posture information include the ankles and shoulders of the human body; collect the position information of the key points of the human posture information of the test person in real time, and analyze the human posture information according to the position information of the key points of the human posture information to determine whether the test person is located in the test area. When the key points of the human body of the test person are all located in the test area, it is determined that the test person is located in the test area; The method for real-time acquisition of the position information of the key points of the human body posture information includes taking a corner in the test area as the origin, the horizontal direction as the x-axis, and the vertical direction as the y-axis to form a Cartesian coordinate system, recording the real-time (x, y1) coordinates of the ankle to complete the acquisition of the ankle position information; recording the real-time y2 coordinates of the shoulder to complete the acquisition of the shoulder position information; S3: During the test time, multiple different y2 coordinates of the shoulders of the tester in a single rope skipping in the test area are collected, and the maximum value of the y2 coordinate of the shoulders collected by the tester in a single rope skipping in the test area is extracted. Based on the y3 coordinate of the average position, the maximum value of the y2 coordinate of the shoulders in multiple rope skipping is compared with the average value y3 coordinate of the rope skipping respectively. When the maximum value of the y2 coordinate of the shoulders in the rope skipping action exceeds 45%-55% of the y3 coordinate of the average position, the rope skipping count is increased by 1; S4: After the tester confirms that the rope skipping is completed, the number of rope skipping is calculated and output.

2. The method for automatically counting rope skipping based on artificial intelligence according to claim 1, characterized in that: In step S3, the method for determining the average position includes: S31: forming a set of calculation intervals according to multiple y2 coordinates of the shoulder position collected by the tester in a single rope skipping, and calculating the median of the calculation interval according to the maximum value and the minimum value in the calculation interval; S32: Calculate the average value based on the median value obtained by the tester's multiple rope skipping, and the obtained average value is the y3 coordinate of the average position.

3. The method for automatically counting rope skipping based on artificial intelligence according to claim 2, characterized in that: The automatic rope skipping counting method based on artificial intelligence also includes the following steps: S5: The duration, number of jumps and y`3 coordinate of the historical average position of each skipping test are stored in the personal file of the corresponding tester, and the skipping speed of the tester is analyzed according to the duration and number of jumps of a single skipping test.

4. The method for automatically counting rope skipping based on artificial intelligence according to claim 3 is characterized in that: In step S3, the method for determining the average position includes: S31: forming a set of calculation intervals according to multiple y2 coordinates of the shoulder position collected by the tester in a single rope skipping, and calculating the median of the calculation interval according to the maximum value and the minimum value in the calculation interval; S32: Calculate the average value based on the median value obtained by the tester's multiple rope skipping and the y'3 coordinate of the historical average position stored in the tester's file, and the obtained average value is the y3 coordinate of the current average position.

5. The method for automatic rope skipping counting based on artificial intelligence according to claim 1, characterized in that: In step S2, after the person to be tested is determined based on the captured facial information, the video interface displayed on the display device screen is cut so that only the key points of all the human body posture information of the test person appear in the video interface.

6. The method for automatic rope skipping counting based on artificial intelligence according to claim 1, characterized in that: In step S2, when it is determined that the test person is located in the test area, rope skipping test start information is output.

7. The method for automatic rope skipping counting based on artificial intelligence according to claim 6 is characterized in that: In step S2, when it is determined that the tester is in the test area, the countdown information for the start of the rope skipping test is inserted in the form of a pop-up window in the screen of the display device. When the countdown to the start of the rope skipping test ends, the pop-up window is closed and the rope skipping actions of the tester begin to be counted.

8. The method for automatic rope skipping counting based on artificial intelligence according to claim 6 is characterized in that: In step S2, when it is determined that the test person is located in the test area, after the voice information of the start of the rope skipping test is output through the voice output device, the rope skipping actions of the test person are counted.