A skipping rope counting method and system based on image recognition

Abstract

The invention relates to the technical field of rope skipping counting, and provides a rope skipping counting method and system based on image recognition. The method comprises: analyzing and identifying historical rope skipping videos, identifying the rotation reference angle of a person's hand in an image corresponding to a rope skipping counting time point; in a subsequent rope skipping video, performing image analysis on video frames between two adjacent rope skipping counting time points, distinguishing the actual rotation angle of the person's hand in the image from the angle that should be rotated to at the corresponding time point based on a recognition model, determining that one rope skipping is correctly completed between the two rope skipping counting time points when they meet, and counting, accurately and efficiently identifying whether the rope jumper is actually skipping, reducing the cost of manual supervision and judgment, and improving the efficiency of rope skipping counting.

Description

Technical Field

[0001] The present invention relates to the technical field of rope skipping counting, and in particular to a rope skipping counting method and system based on image recognition. Background Art

[0002] In the prior art, manual referees are generally used to supervise rope skipping, and manual counting is combined with machine counting to ensure that the results are reliable and accurate. However, manual referees not only consume energy in the process of continuous supervision and counting, but are also prone to missed calls. If electronic referees are used, it is difficult to distinguish the rope speed during rope skipping, and rope jumpers may cheat by faking jumps. Rope skipping counting is inefficient and difficult to count. Summary of the invention

[0003] The invention provides a rope skipping counting method based on image recognition, which is used to solve the problems of low efficiency and difficulty in rope skipping counting in the prior art.

[0004] A first aspect of the present invention provides a method for counting rope skipping based on image recognition, comprising:

[0005] Obtain historical rope skipping videos and rope skipping counting time points, and obtain the first rope skipping image of the corresponding frame according to the rope skipping counting time points; perform image processing on the first rope skipping image to identify the reference angle of rotation of the human hand;

[0006] Receive the counting time point corresponding to the skipping rope counting signal, and obtain the current skipping rope video; randomly select a moment in the time period formed by two adjacent counting time points as the recognition time point, and use the frame corresponding to the recognition time point in the current skipping rope video as the second skipping rope image, and identify the actual rotation angle of the human hand in the second skipping rope image; substitute each time point and the actual rotation angle of the human hand into a preset recognition model to determine whether the time point satisfies the recognition model, and the recognition model is:

[0007]

[0008] Wherein, T is the recognition time point of the second rope skipping image, t1 is the smaller value among the rope skipping counting time points, t2 is the larger value among the rope skipping counting time points, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, δ is the correction angle value;

[0009] If the identification model is satisfied, one rope skipping attempt is counted.

[0010] Optionally, before counting the number of rope skipping as one if the identification model is satisfied, the method further includes:

[0011] If the identification model is not satisfied, it is determined whether the current identification time point is the first or last circle of rope skipping. If not, it is determined to be a false jump and the current rope skipping count is ignored.

[0012] Optionally, the reference angle of hand rotation of the identified person is specifically:

[0013] Identify the wrist feature points and finger feature points of the person in the first rope skipping image, connect the wrist feature points and finger feature points to obtain a reference direction, establish a coordinate system with the wrist feature points and identify the angle of the reference direction to obtain a reference angle of hand rotation.

[0014] The second aspect of the present application provides a skipping rope counting system based on image recognition, comprising:

[0015] The reference angle recognition module is used to obtain the historical rope skipping video and the rope skipping counting time point, and obtain the first rope skipping image of the corresponding frame according to the rope skipping counting time point; perform image processing on the first rope skipping image to identify the reference angle of human hand rotation;

[0016] The angle recognition module is used to receive the counting time point corresponding to the skipping rope counting signal and obtain the current skipping rope video; randomly select a moment in the time period formed by two adjacent counting time points as the recognition time point, and use the frame corresponding to the recognition time point in the current skipping rope video as the second skipping rope image to identify the actual rotation angle of the human hand in the second skipping rope image; substitute each time point and the actual rotation angle of the human hand into a preset recognition model to determine whether the time point satisfies the recognition model, and the recognition model is:

[0017]

[0018] Wherein, T is the recognition time point of the second rope skipping image, t1 is the smaller value among the rope skipping counting time points, t2 is the larger value among the rope skipping counting time points, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, δ is the correction angle value;

[0019] The rope skipping counting module is used to count the number of rope skipping as one if the identification model is met.

[0020] Optionally, in the rope skipping counting module, if the identification model is satisfied, before counting the number of rope skipping times, the module further includes:

[0021] If the identification model is not satisfied, it is determined whether the current identification time point is the first or last circle of rope skipping. If not, it is determined to be a false jump and the current rope skipping count is ignored.

[0022] Optionally, in the reference angle recognition module, the reference angle of hand rotation is specifically identified as:

[0023] Identify the wrist feature points and finger feature points of the person in the first rope skipping image, connect the wrist feature points and finger feature points to obtain a reference direction, establish a coordinate system with the wrist feature points and identify the angle of the reference direction to obtain a reference angle of hand rotation.

[0024] The third aspect of the present application provides a method and device for counting rope skipping based on image recognition, the device comprising a processor and a memory:

[0025] The memory is used to store program code and transmit the program code to the processor;

[0026] The processor is used to execute the rope skipping counting method based on image recognition according to the instructions in the program code.

[0027] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store program code, and the program code is used to execute a skipping rope counting method based on image recognition as described in any one of the first aspects of the present invention.

[0028] It can be seen from the above technical scheme that the present invention has the following advantages: by analyzing and identifying historical skipping rope videos, the reference rotation angle of the human hand in the image corresponding to the skipping rope counting time point is identified; in subsequent skipping rope videos, the video frames between two adjacent skipping rope counting time points are subjected to image analysis, and the actual rotation angle of the human hand in the image is distinguished from the angle that should be rotated at the corresponding time point based on the recognition model. When they meet the requirements, it is determined that one skipping rope is correctly completed between the two skipping rope counting time points, and the number of times is counted, so as to accurately and efficiently identify whether the jumper is actually skipping rope, reduce the cost of manual supervision and judgment, and improve the efficiency of skipping rope counting. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 is a flow chart of a method for counting rope skipping based on image recognition;

[0031] Figure 2 This is a structural diagram of a rope skipping counting system based on image recognition. DETAILED DESCRIPTION

[0032] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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.

[0033] The invention provides a rope skipping counting method based on image recognition, which is used to solve the problems of low efficiency and difficulty in rope skipping counting in the prior art.

[0034] See also Figure 1 , Figure 1 The present invention provides a first flow chart of a method for counting rope skipping based on image recognition according to an embodiment of the present invention.

[0035] S100, obtaining a historical rope skipping video and a rope skipping counting time point, and obtaining a first rope skipping image of a corresponding frame according to the rope skipping counting time point; performing image processing on the first rope skipping image to identify a reference angle of rotation of a human hand;

[0036] It should be noted that when the current rope skipping user performs multiple rope skipping movements, the completed rope skipping movements should be filmed as a historical rope skipping video, and the shooting direction of the video should be consistent with the direction of the rotation axis of the rope skipping as much as possible, that is, it should be filmed from the side of the rope jumper, so that the video can better record the rotation of the human hand; for example, when a one-minute rope skipping exercise is required, the first few seconds of the rope skipping exercise will be used as the historical rope skipping video, and the subsequent rope skipping video of the remaining time will be used to judge the subsequent steps. This setting method can ensure that the human hand characteristics obtained at the counting time points throughout the rope skipping process are consistent. The counting mechanism principle of the counting rope skipping can adopt a mechanical gear structure, a photoelectric sensor structure, a magnetic field sensor structure or a gravity sensor structure. The counting rope skipping will count at a specific position when the rope passes around the handle, and in a continuous The handle holding position will basically not change during the rope skipping exercise, so the human hand rotation reference angle obtained from the previously completed historical rope skipping video can be used as the subsequent judgment standard; if the handle holding position is limited during the rope skipping exercise, the completed historical video of the non-continuous rope skipping exercise can be used for image processing to identify the human hand rotation reference angle. When the handle holding position is limited, it can be ensured that the handle holding position of all rope skipping exercises is consistent. For example, the handle has a hand-shaped depression and the jumper needs to hold it in a fixed direction, or there is a requirement that the LCD screen should face up during rope skipping; if there are multiple jumpers in the rope skipping video at the same time, it is necessary to plan the image area in advance, associate each image area with the rope skipping counting signal, obtain the corresponding frame image when the rope skipping counting point signal corresponding to the different area is received, and perform image segmentation in the corresponding area.

[0037] While obtaining the historical skipping rope video, the skipping rope counting signal detected by the skipping rope counter will also be received. Each time a signal is received, the user's completion of a skipping rope and the corresponding time point will be recorded to obtain the corresponding skipping rope counting time point; by comparing the skipping rope counting time point with the timeline of the historical skipping rope video, the first skipping rope image in the frame corresponding to the time point is obtained; the first skipping rope image can be subjected to image denoising, and human body recognition can be performed to segment the human body area image, and the hand orientation features of the human body area image can be identified based on a preset neural network model to obtain a reference angle for human hand rotation. The preset neural network model can be pre-trained based on some skipping rope pictures to identify the human hand's wrist, set a coordinate axis at the wrist, and identify the line connecting the fingertip joint of the human hand and the wrist, and use the angle of the line on the coordinate axis as the reference angle for human hand rotation, which corresponds to the skipping rope counting time point; the angle position in the subsequent counting is only used to determine the relative position relationship, so in this embodiment, it can be set to 0° in either direction of the wrist coordinate axis.

[0038] S200, receiving a counting time point corresponding to a skipping rope counting signal, and obtaining a current skipping rope video; randomly selecting a moment in a time period formed by two adjacent counting time points as a recognition time point, and using a frame corresponding to the recognition time point in the current skipping rope video as a second skipping rope image, and identifying an actual rotation angle of a person's hand in the second skipping rope image; substituting each time point and the actual rotation angle of a person's hand into a preset recognition model, and determining whether the time point satisfies the recognition model, wherein the recognition model is:

[0039]

[0040] Wherein, T is the recognition time point of the second rope skipping image, t1 is the smaller value among the rope skipping counting time points, t2 is the larger value among the rope skipping counting time points, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, δ is the correction angle value;

[0041] It should be noted that in the state of continuous rope skipping, the jumper's hands and limbs move in coordination. The jumper will keep the speed of each circle of the rope uniform to maintain the stability of the rope skipping. In addition to the acceleration when taking off and the deceleration before stopping, the speed of each circle of the rope skipping at each angle should be uniform, so as to maintain the rhythm of rope skipping. The rope skipping rotates with the rotation of the human hand, so the angle of the human hand in each circle should correspond to the position of the rope skipping. The actual angle of the human hand in the recognition image can be compared with the corresponding angle at the corresponding moment to determine whether there is real rope skipping;

[0042] In this embodiment, the skipping rope counting signal is received in real time. Each time the skipping rope is counted, the counting signal received after passing the counting trigger point has a corresponding counting time point, and the current skipping rope video will be shot at the same time, and the counting time point is consistent with the timeline of the current skipping rope video; between every two adjacent skipping rope counting time points, the jumper is deemed to have completed one round of skipping rope, and any moment in the time period formed between the two counting time points is taken as the identification time point, and the image of the corresponding frame in the current skipping rope video is taken as the second skipping rope image. At this time, the actual angle of the person's hand rotation in the second skipping rope image should correspond to the position of the identification time point in the time period formed between the two counting time points; the random selection of the identification time point can well avoid the contingency of the person's hand angle being consistent with the response angle during false jumping, thereby reducing the error of subsequent judgment.

[0043] Through the identification model, the difference between the larger value and the smaller value in the skipping rope counting time point can be used to obtain the length of the time period, which varies from person to person. The position of the identification time point in the time period can be obtained by subtracting the smaller value of the identification time point from the skipping rope counting time point and dividing it by the length of the time period. Multiplying it by 360° of a skipping rope circle plus the reference angle can obtain the human hand angle that should correspond to the identification time point. Considering that although each circle of the skipping rope can be regarded as a uniform speed as a whole, it is actually not uniform and there is a certain error, when judging whether the actual angle of the human hand rotation should correspond to the identification time point, the correction angle can be added, and the corresponding angle value can be expanded to a range. Within this range, it is regarded as the correct corresponding skipping rope, and the skipping rope within this time period is a real jump. The correction angle value is set according to the experience value.

[0044] S300: If the identification model is satisfied, then one rope skipping is counted.

[0045] It should be noted that when the identification model of the aforementioned step S200 is satisfied, it can be determined that the jumper has completed one jump between two counting time points, which can be counted as one jump, and the larger value of the jump counting time point in the aforementioned step S200 is retained, and the next adjacent counting time point in the jump counting signal is selected to form another time period again, and the identification time point is selected, and the identification of the aforementioned step S200 is continued until all the jump counting time points are judged.

[0046] In the present embodiment, historical rope skipping videos are analyzed and identified to identify the reference rotation angle of the human hand in the image corresponding to the rope skipping counting time point; in subsequent rope skipping videos, image analysis is performed on the video frames between two adjacent rope skipping counting time points, and the actual rotation angle of the human hand in the image is distinguished from the angle that should be rotated at the corresponding time point based on the recognition model. When they meet the requirements, it is determined that one rope skipping is correctly completed between the two rope skipping counting time points, and the number of times is counted, so as to accurately and efficiently identify whether the jumper is actually skipping, reduce the cost of manual supervision and judgment, and improve the efficiency of rope skipping counting.

[0047] The above is a detailed description of the first embodiment of a rope skipping counting method based on image recognition provided by the present application. The following is a detailed description of the second embodiment of a rope skipping counting method based on image recognition provided by the present application.

[0048] In the present embodiment, a skipping rope counting method based on image recognition is further provided, and the aforementioned step S300 also includes: if the recognition model is not satisfied, then determining whether the current recognition time point is the first or last circle of the skipping rope, if otherwise determining that the current is a false jump, and ignoring the current skipping rope count; it should be noted that, when the recognition model is not satisfied, it is also possible that the skipping rope in this circle does not maintain a uniform speed, and generally, non-uniform speed will occur in the acceleration phase of the first circle of the skipping rope at the beginning or the deceleration phase of the last circle of the skipping rope at the end. Therefore, it can be determined whether the smaller value of the counting time point corresponding to the current recognition time point is the minimum value among the counting time points, and whether the larger value of the counting time point is the maximum value among the counting time points, to determine whether it is the first or last circle, and determine whether the first or last circle is included in the counting of the number of circles according to the actual skipping rope rules; and if it is not the first or last circle, it can be determined that the circle corresponding to the current recognition time point is a false jump, and the number of circles should not be included in the counting.

[0049] Furthermore, the identification of the reference angle of human hand rotation described in the aforementioned step S100 specifically includes: identifying the human wrist feature point and the finger feature point in the first skipping rope image, connecting the human wrist feature point and the finger feature point to obtain a reference direction, establishing a coordinate system with the human wrist feature point and identifying the angle of the reference direction, to obtain the human hand rotation reference angle; it should be noted that generally a person's hand will clamp the skipping rope handle between the tip of the thumb and the second joint of the index finger, so the finger feature point can be a point at the tip of the thumb or the second joint of the index finger, and the rotation of the wrist during skipping drives the skipping rope to rotate, then the rotation angle of the skipping rope corresponds to the rotation angle of the finger feature point around the wrist feature point.

[0050] The above is a detailed description of a rope skipping counting method based on image recognition according to the first aspect of the present application. The following is a detailed description of an embodiment of a rope skipping counting system based on image recognition according to the second aspect of the present application.

[0051] See also Figure 2 , Figure 2 This embodiment provides a rope skipping counting system based on image recognition, including:

[0052] The reference angle recognition module 10 is used to obtain the historical rope skipping video and the rope skipping counting time point, and obtain the first rope skipping image of the corresponding frame according to the rope skipping counting time point; perform image processing on the first rope skipping image to identify the reference angle of the human hand rotation;

[0053] The angle recognition module 20 is used to receive the counting time point corresponding to the skipping rope counting signal and obtain the current skipping rope video; randomly select a moment in the time period formed by two adjacent counting time points as the recognition time point, and use the frame corresponding to the recognition time point in the current skipping rope video as the second skipping rope image to identify the actual rotation angle of the human hand in the second skipping rope image; substitute each time point and the actual rotation angle of the human hand into a preset recognition model to determine whether the time point satisfies the recognition model, and the recognition model is:

[0054]

[0055] Wherein, T is the recognition time point of the second rope skipping image, t1 is the smaller value among the rope skipping counting time points, t2 is the larger value among the rope skipping counting time points, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, δ is the correction angle value;

[0056] The rope skipping counting module 30 is used to count the number of rope skipping as one if the identification model is satisfied.

[0057] Furthermore, in the rope skipping counting module 30, if the identification model is satisfied, before counting the number of rope skipping times, the following is further included:

[0058] If the identification model is not satisfied, it is determined whether the current identification time point is the first or last circle of rope skipping. If not, it is determined to be a false jump and the current rope skipping count is ignored.

[0059] Furthermore, in the reference angle recognition module 10, the reference angle of hand rotation is specifically identified as:

[0060] Identify the wrist feature points and finger feature points of the person in the first rope skipping image, connect the wrist feature points and finger feature points to obtain a reference direction, establish a coordinate system with the wrist feature points and identify the angle of the reference direction to obtain a reference angle of hand rotation.

[0061] The third aspect of the present application also provides a skipping rope counting method device based on image recognition, including a processor and a memory: the memory is used to store program code and transmit the program code to the processor; the processor is used to execute the above-mentioned skipping rope counting method based on image recognition according to the instructions in the program code.

[0062] A fourth aspect of the present application provides a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store program code, and the program code is used to execute the above-mentioned image recognition-based rope skipping counting method.

[0063] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and equipment can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0064] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0065] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0066] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0067] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0068] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for counting rope skipping based on image recognition, characterized in that include: Obtain historical rope skipping videos and rope skipping counting time points, wherein the rope skipping counting time points are time points at which the rope skipping passes through a specific counting position and sends out a rope skipping counting signal, and obtain a first rope skipping image of a corresponding frame according to the rope skipping counting time points; perform image processing on the first rope skipping image to identify a reference angle of rotation of a person's hand; Receive the counting time point corresponding to the skipping rope counting signal, and obtain the current skipping rope video; randomly select a moment in the time period formed by two adjacent counting time points as the recognition time point, and use the frame corresponding to the recognition time point in the current skipping rope video as the second skipping rope image, and identify the actual rotation angle of the human hand in the second skipping rope image; substitute each time point and the actual rotation angle of the human hand into a preset recognition model to determine whether the time point satisfies the recognition model, and the recognition model is: Wherein, T is the recognition time point of the second rope skipping image, t1 is the initial value of the rope skipping counting time period, t2 is the end value of the rope skipping counting time period, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, and δ is the correction angle value; If the identification model is satisfied, one rope skipping attempt is counted.

2. The method for counting rope skipping based on image recognition according to claim 1, characterized in that: If the identification model is satisfied, before counting the number of rope skipping, the following is also included: If the identification model is not satisfied, it is determined whether the current identification time point is the first or last circle of rope skipping. If not, it is determined that the current jump is a false jump and the current rope skipping count is ignored.

3. The method for counting rope skipping based on image recognition according to claim 1, characterized in that: The reference angle for identifying the rotation of a person's hand is specifically: Identify the wrist feature points and finger feature points of the person in the first rope skipping image, connect the wrist feature points and finger feature points to obtain a reference direction, establish a coordinate system with the wrist feature points and identify the angle of the reference direction to obtain a reference angle of hand rotation.

4. A skipping rope counting system based on image recognition, characterized in that: include: The reference angle recognition module is used to obtain historical rope skipping videos and rope skipping counting time points, wherein the rope skipping counting time points are the time points at which the rope skipping passes through a specific counting position and sends out a rope skipping counting signal, and obtain the first rope skipping image of the corresponding frame according to the rope skipping counting time points; perform image processing on the first rope skipping image to identify the reference angle of rotation of the human hand; The angle recognition module is used to receive the counting time point corresponding to the skipping rope counting signal and obtain the current skipping rope video; randomly select a moment in the time period formed by two adjacent counting time points as the recognition time point, and use the frame corresponding to the recognition time point in the current skipping rope video as the second skipping rope image to identify the actual rotation angle of the human hand in the second skipping rope image; substitute each time point and the actual rotation angle of the human hand into a preset recognition model to determine whether the time point satisfies the recognition model, and the recognition model is: Wherein, T is the recognition time point of the second rope skipping image, t1 is the initial value of the rope skipping counting time period, t2 is the end value of the rope skipping counting time period, θ is the reference angle of human hand rotation, is the actual angle of human hand rotation, and δ is the correction angle value; The rope skipping counting module is used to count the number of rope skipping as one if the identification model is met.

5. The image recognition-based skipping rope counting system according to claim 4, characterized in that: In the rope skipping counting module, if the identification model is satisfied, before counting the number of rope skipping times, the following is also included: If the identification model is not satisfied, it is determined whether the current identification time point is the first or last circle of rope skipping. If not, it is determined that the current jump is a false jump and the current rope skipping count is ignored.

6. The image recognition-based skipping rope counting system according to claim 4, characterized in that: In the reference angle recognition module, the reference angle for identifying the rotation of a person's hand is specifically: Identify the wrist feature points and finger feature points of the person in the first rope skipping image, connect the wrist feature points and finger feature points to obtain a reference direction, establish a coordinate system with the wrist feature points and identify the angle of the reference direction to obtain a reference angle of hand rotation.

7. A skipping rope counting device based on image recognition, characterized in that: The device comprises a processor and a memory: The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the rope skipping counting method based on image recognition as described in any one of claims 1 to 3 according to the instructions in the program code.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and the program codes are used to execute the rope skipping counting method based on image recognition as described in any one of claims 1 to 3.

Images