In-situ longitudinal jump action evaluation method and device

Through the feature analysis of the action image sequence, the normativeness and stability of the in-situ vertical jump action are evaluated, and the problems of high computing power and single evaluation dimensions in the prior art are solved, and real-time and accurate multi-dimensional evaluation results are achieved.

CN118976232AActive Publication Date: 2024-11-19NANJING INST OF TECH
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202410978497.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-19
Estimated Expiration
2044-07-22

AI Technical Summary

Technical Problem

The existing technology has high requirements for equipment computing power, complex identification of human skeleton points, which can easily lead to incorrect evaluation, and the evaluation dimension is single and lack reference.

Method used

By obtaining the action image sequence, selecting characteristic action images to determine the position of the bone point, determining whether the jump, jump and landing movements comply with the specifications, combining the three-dimensional coordinate system to evaluate the trunk swing and hip and knee rotation, providing multi-dimensional action evaluation.

Benefits of technology

Real-time and accurate in-situ vertical jump action evaluation is achieved, reducing the computing power requirements of the equipment, providing multi-dimensional evaluation results, and is of reference value.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118976232A_ABST
    Figure CN118976232A_ABST
Patent Text Reader

Abstract

The invention discloses an in-situ longitudinal jump action evaluation method and device, and the method comprises the steps: a tester executes an in-situ longitudinal jump action, and obtains an action image sequence in the action process of the tester; judging whether the postures and the positions of the two arms and the two legs of the tester conform to take-off specifications or not; judging whether the trunk swing amplitude of the tester conforms to a longitudinal jump specification or not; judging whether the rotation amplitudes of the hip and knee joints and the feet of the testers conform to the non-motion compensation specification or not; and a judgment result of action evaluation of the tester is provided. By the adoption of the technical scheme, real-time performance is achieved, testing personnel can conduct testing at any time, the requirement for computing power of equipment is not high, the longitudinal jumping action is evaluated in a multi-dimensional mode by setting universal longitudinal jumping body posture and position standards, the evaluation result is accurate, and reference value is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of sports ability assessment, and in particular to a method and device for assessing vertical jump in situ. Background Art

[0002] Vertical jumping in place is a common aerobic exercise and one of the commonly used training movements in many fitness trainings. It can effectively exercise the lower limb muscles, improve cardiopulmonary function, enhance physical fitness and endurance. It does not require any equipment and can be performed anytime and anywhere. It is a very convenient and effective way of exercise. However, the key to a good vertical jumping in place posture lies in the trainer's jumping height, stability, and the judgment of whether there are any movement compensation items during the vertical jumping in place. If the vertical jumping in place is not standardized or the exercise is overloaded, it may cause strain on the body's bones and joints, especially the imbalanced state and excess movement compensation during the exercise will cause serious damage to the leg muscles and ankles. Therefore, it is particularly important to standardize the movement during the vertical jumping in place.

[0003] Most of the existing evaluation schemes for vertical jump in place are carried out through the recognition method of neural network models. The problems are that the recognition of human skeleton points is pre-processed, which fails to allow users to use it as needed. In addition, the use of neural networks to process video input to obtain the coordinate information of human skeleton points requires extremely high equipment computing power and a complex processing method. The recognition and classification of human features are also limited by the features contained in the training parameters of the neural network model, which is prone to incorrect evaluation. In addition, there is a lack of evaluation of typified postures during vertical jumps (vertical jump stability, motion compensation, etc.), and the evaluation dimension is single, resulting in a lack of reference for the evaluation results. Summary of the invention

[0004] Purpose of the invention: The present invention provides a method and device for evaluating vertical jump in situ, aiming to solve the technical problems in the prior art of extremely high requirements on equipment computing power, the need for preprocessing of human skeleton point recognition, the tendency to make incorrect evaluations due to the characteristics implied by the training parameters of the neural network model, and the single evaluation dimension for vertical jumps that lacks reference.

[0005] Technical solution: The present invention provides a method for evaluating an in-situ vertical jump action, comprising: a test person performs an in-situ vertical jump action, and obtains a sequence of action images of the test person during the action; selects a characteristic action image from the action image sequence, and determines the position of the test person's skeleton points in the action image; judges whether the posture and position of the test person's arms and legs meet the take-off specifications through the positions of the skeleton points in the test person's take-off action in the characteristic action image, and if they meet the take-off specifications, determines that the test person's in-situ vertical jump is valid; judges whether the test person's torso swing amplitude meets the vertical jump specifications through the positions of the skeleton points in the test person's jumping action in the characteristic action image; judges whether the rotation amplitude of the test person's hip, knee and foot meets the no-motion compensation specifications through the positions of the skeleton points in the test person's take-off action and landing action in the characteristic action image; and provides a judgment result of the test person's action evaluation.

[0006] Specifically, the bone point positions include: central lumbar point, left shoulder, right shoulder, left heel, right heel, left toe, right toe, left elbow joint, right elbow joint, left knee joint, right knee joint, left hip joint, right hip joint, throat, and crotch.

[0007] Specifically, an action image in which the test person is at the highest height from the ground is selected from the action image sequence, and the shooting time of the action image with the highest height from the ground is used as the first reference time, and the jumping action images before and after the first reference time are selected according to a predetermined time interval; an action image in which the test person squats the lowest is selected from the action image sequence, and the shooting time of the action image in which the test person squats the lowest is used as the second reference time, and the taking-off action images before and after the second reference time are selected according to a predetermined time interval; an action image in which the test person lands and touches the ground is selected from the action image sequence, and the shooting time of the action image in which the test person lands and touches the ground is used as the third reference time, and the landing action image after the third reference time is selected according to a predetermined time interval; the selected jumping action image, taking-off action image and landing action image are used as feature action images.

[0008] Specifically, the take-off specifications include: the distance from the left heel to the central lumbar point and the distance from the right heel to the central lumbar point, the distance deviation is less than a first distance threshold; the distance from the left heel to the right heel is greater than the distance from the left shoulder to the right shoulder, and the distance difference is less than a second distance threshold; the distance from the left shoulder to the left elbow joint and the distance from the right shoulder to the right elbow joint are both greater than the swing distance threshold; the angle of the left elbow joint and the angle of the right elbow joint are both less than the first angle threshold; the angle of the left knee joint and the angle of the right knee joint are both less than the second angle threshold.

[0009] Specifically, the positions of the skeletal points in the test person's jumping action in the feature action image are used to determine whether the test person's vertical jump height meets the jumping specifications; the positions of the skeletal points in the test person's landing action in the feature action image are used to determine whether the test person's knee flexion action meets the landing specifications; the landing specifications include: the angle of the left knee joint and the angle of the right knee joint, both of which are less than the third angle threshold; if the take-off specifications, jumping specifications and landing specifications are met at the same time, the test person's vertical jump in place is deemed valid.

[0010] Specifically, the vertical jump specifications include: the x-axis distance from the left shoulder to the vertical line where the center lumbar point is located, and the x-axis distance from the right shoulder to the vertical line where the center lumbar point is located, and the x-axis distance is less than the third distance threshold; the y-axis distance from the left shoulder to the vertical line where the center lumbar point is located, and the y-axis distance from the right shoulder to the vertical line where the center lumbar point is located, and the y-axis distance is less than the fourth distance threshold; the vertical swing angle formed by the throat and the crotch and the center lumbar point respectively, and the vertical swing angle change is less than the fourth angle threshold; with the center lumbar point as the coordinate origin, the direction of the tester's shoulders as the x-axis, the direction from the heel to the toes as the y-axis, and the height direction as the z-axis, a three-dimensional coordinate system is established.

[0011] Specifically, the no-motion compensation norms include: a first hip-knee rotation angle formed by a line connecting the left hip joint, the left knee joint and the central lumbar point, a second hip-knee rotation angle formed by a line connecting the right hip joint, the right knee joint and the central lumbar point, a change in the first hip-knee rotation angle and the second hip-knee rotation angle is less than a fifth angle threshold; a first y-axis distance between the left heel and the left toe, a second y-axis distance between the right heel and the right toe, and a change in the first y-axis distance and the second y-axis distance is less than a fifth distance threshold.

[0012] Specifically, under the premise that the vertical jump in place of the test person is valid, if both the vertical jump standard and the no-motion compensation standard are met, the test person's action is judged to be a standard vertical jump in place action.

[0013] Specifically, the corresponding height score is calculated according to the vertical jump height, the corresponding stability score is calculated according to the trunk swing amplitude, and the corresponding non-motion compensation score is calculated according to the rotation amplitude of the hip, knee and foot. The height score, stability score and non-motion compensation score are weighted and added together to calculate the quality score of the in-situ vertical jump action.

[0014] The present invention provides an in-situ vertical jump action evaluation device, comprising: an image acquisition unit, a skeleton point position determination unit, a take-off action evaluation unit, a jumping action evaluation unit, a motion compensation evaluation unit and a comprehensive evaluation unit, wherein: the image acquisition unit is used to test a person performing an in-situ vertical jump action and obtain a sequence of motion images of the person performing the action; the skeleton point position determination unit is used to select a feature motion image from the motion image sequence and determine the position of the skeleton point of the person performing the action in the motion image; the take-off action evaluation unit is used to determine the test person's take-off action through the skeleton point position of the person performing the take-off action in the feature motion image. Whether the posture and position of the test person's arms and legs meet the take-off specifications. If they meet the take-off specifications, the test person's vertical jump in place is deemed valid; the jumping action evaluation unit is used to judge whether the test person's torso swing amplitude meets the vertical jump specifications through the positions of the skeletal points in the test person's jumping action in the characteristic action image; the motion compensation evaluation unit is used to judge whether the rotation amplitude of the test person's hip, knee and foot meets the no-motion compensation specifications through the positions of the skeletal points in the test person's take-off and landing actions in the characteristic action image; the comprehensive evaluation unit is used to provide the judgment result of the test person's action evaluation.

[0015] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: it is real-time, testers can do the test whenever they want, and the computing power of the equipment is not high. By setting universal vertical jump body posture and position standards, the vertical jump action is evaluated in multiple dimensions, and the evaluation result is relatively accurate and has reference value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the structure of a vertical jump action evaluation device provided by the present invention;

[0017] Figure 2 A schematic flow chart of the vertical jump action evaluation method provided by the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0019] See also Figure 1 , which is a structural schematic diagram of the vertical jump action evaluation device provided by the present invention.

[0020] In a specific implementation, it includes a Microsoft Xbox 360 body camera 1 on the front side, a Microsoft Xbox 360 body camera 2 on the other side, and a computer device 3 connected to the camera.

[0021] In the specific implementation, it should be noted that the Microsoft Xbox360 kinect used in the present invention is a 3D somatosensory camera with a somatosensory peripheral peripheral, which has a color and depth of field lens, a voice microphone array, and a motorized base for automatically adjusting the angle of the Kinect body. The viewing angle is 57 degrees horizontally, 43 degrees vertically, and the physical tilt range is ±27 degrees. The depth of field lens sensing distance is 1.2 meters to 3.5 meters; it can capture the skeleton tracking system of 2 people at the same time. It relies on the camera to capture the user's movement in three-dimensional space, and also has multi-language voice recognition support, which meets the use requirements of the present invention.

[0022] In the specific implementation, in this embodiment, the computer device 3 connected to the camera is specifically Lenovo Legion Y9000P 2023 (32GB / 2TB), and the CPU model is Intel i9-13900HX. It has 8 performance cores (P-Cores) and 16 energy efficiency cores (E-Cores), providing a total of 24 threads. Its computing speed and storage space can meet the requirements of video analysis and modeling of two high-definition cameras, and can generate a human skeleton point coordinate system in real time as required, and perform data analysis and data storage.

[0023] See also Figure 2 , which is a flow chart of the vertical jump action evaluation method provided by the present invention.

[0024] In the embodiment of the present invention, as described in step 1, the test person performs a vertical jump in place, and the image acquisition unit acquires a sequence of action images of the test person during the action.

[0025] In a specific implementation, before the tester enters the specified area, the Kinect body camera is placed in front of and at an appropriate distance to the left to ensure that the human body is within the camera field of view. It should be noted that the angles of the two cameras used in the present invention, which are located in front of and on the left of the user, can be placed according to the specific situation and the angle that the camera can detect.

[0026] In the specific implementation, after completing the identity authentication or identity login, the Kinect body camera begins to identify the test person's bone point coordinate information and key data, including shoulder width, upper arm length, forearm length, shoulder-vertebrae distance, hip length, thigh length, calf length and foot distance.

[0027] In the specific implementation, it should be noted that the user should prepare after standing in the specified area, stand in a straddle position, make sure the distance between the feet is slightly wider than the shoulders, bend the knees and swing the arms down to prepare for the jump (refer to the specific requirements of the jump specifications). At the same time, the computer analyzes the data transmitted by the body camera to see whether the user's preparation posture is correct, and gives a reminder if the preparation action is incorrect.

[0028] In the embodiment of the present invention, as described in step 2, a characteristic action image is selected from the action image sequence, and the skeleton point position determination unit determines the skeleton point position of the test person in the action image.

[0029] In the embodiment of the present invention, the positions of the bone points include: central lumbar point, left shoulder, right shoulder, left heel, right heel, left toe, right toe, left elbow joint, right elbow joint, left knee joint, right knee joint, left hip joint, right hip joint, throat, and crotch.

[0030] In a specific implementation, the coordinate position of the left shoulder is A(X a ,Y a ,Z a ), the coordinate position of the right shoulder is B(X b ,Y b ,Z b ), the coordinate position of the left elbow joint is C(X c ,Y c ,Z c ), the coordinate position of the right elbow joint is D(X d ,Y d ,Z d ), the coordinate position of the central lumbar point (the center point of the lumbar spine) is E(X e ,Y e ,Z e ), the coordinate position of the left hip joint is F(X f ,Y f ,Z f ), the coordinate position of the right hip joint is G(X g ,Y g ,Z g ), the coordinate position of the left knee is H(X h ,Y h ,Z h ), the coordinate position of the right knee is I(X i ,Y i ,Z i ), the coordinate position of the left heel is J(X j ,Y j ,Z j ), the coordinate position of the right heel is K(X k ,Y k ,Z k ), the coordinate position of the throat point is M(X m ,Y m ,Z m ), the coordinate position of the crotch point is N(X n ,Y n ,Z n ), the coordinate position of the left toe point is O(X o ,Y o ,Zo ), the coordinate position of the right toe point is P(X p ,Y p ,Z p ).

[0031] In an embodiment of the present invention, an action image in which the test person has the highest height above the ground is selected from the action image sequence, and the shooting time of the action image with the highest height above the ground is used as the first reference time, and jumping action images before and after the first reference time are selected at a predetermined time interval (which can be set accordingly according to the actual application scenario); an action image in which the test person squats the lowest is selected from the action image sequence, and the shooting time of the action image in which the test person squats the lowest is used as the second reference time, and taking-off action images before and after the second reference time are selected at a predetermined time interval; an action image in which the test person lands (just) on the ground is selected from the action image sequence, and the shooting time of the action image in which the test person lands on the ground is used as the third reference time, and the landing action image after the third reference time is selected at a predetermined time interval; the selected jumping action image, take-off action image and landing action image are used as feature action images.

[0032] In a specific implementation, by selecting multiple jumping action images, take-off action images and landing action images, the body posture of the test person when performing the action and the range of change of the posture can be determined and analyzed, thereby determining whether the corresponding vertical jump specifications are met.

[0033] In the specific implementation, multiple pictures that can represent the body posture of the test person during the vertical jump are selected. Compared with video processing, picture processing can greatly reduce the computing power requirements of the device.

[0034] In the specific implementation, the camera captures the tester's motion image, the computer starts to analyze, and after selecting the tester's jumping quality pictures, the obtained human activity state pictures are processed in real time, and the key bone coordinate points of the user are obtained based on the Kinect body camera. By connecting to the host computer and display equipped with Kinect SDK, the collected coordinate position information reads the bone node data through the interface function provided by the SDK, and then transmits it to the database of the host computer. Finally, the Python language program is used to complete the calculation.

[0035] In an embodiment of the present invention, as described in step 3, the take-off action evaluation unit determines whether the posture and position of the test person's arms and legs meet the take-off specifications through the positions of the skeleton points in the test person's take-off action in the characteristic action image. If they meet the take-off specifications, the test person's in-situ vertical jump is deemed valid.

[0036] In the specific implementation, during the jumping movement, the tester should be required to jump up from the spot, swing the upper arms up, and form an abdominal stretch in the air. When landing, the tester should consciously retract the abdomen and bend the knees. The movement compensation items should include judging whether the hip and knee joints rotate when jumping and touching the ground, and whether the feet rotate when jumping and touching the ground. The key points of the movement can be summarized as: standing in a straddle position, bending the knees and swinging the arms down to prepare for the jump, jumping up from the spot, swinging the upper arms up, and forming an abdominal stretch in the air. When landing, the tester should consciously retract the abdomen, bend the knees to cushion the impact, and swing the arms down to restore the jump position.

[0037] In an embodiment of the present invention, the take-off specification includes: the distance from the left heel to the central lumbar point and the distance from the right heel to the central lumbar point, the distance deviation is less than a first distance threshold; the distance from the left heel to the right heel is greater than the distance from the left shoulder to the right shoulder, and the distance difference is less than a second distance threshold; the distance from the left shoulder to the left elbow joint and the distance from the right shoulder to the right elbow joint are both greater than the swing distance threshold; the angle of the left elbow joint and the angle of the right elbow joint are both less than the first angle threshold; the angle of the left knee joint and the angle of the right knee joint are both less than the second angle threshold.

[0038] In a specific implementation, the deviation between the distance from the left heel to the central lumbar point and the distance from the right heel to the central lumbar point reflects that the two legs remain relatively symmetrical with respect to the center line of the human body.

[0039] In a specific implementation, the distance from the left heel to the right heel is greater than the distance from the left shoulder to the right shoulder, which means that the distance between the two feet should be slightly greater than the distance between the two shoulders.

[0040] In a specific implementation, the distance from the left shoulder to the left elbow joint, and the distance from the right shoulder to the right elbow joint, reflect whether the arms are swinging and the swing amplitude.

[0041] In a specific implementation, the angle of the left elbow joint and the angle of the right elbow joint reflect whether the arms are in a bent state and the degree of bending. The angle of the left and right elbow joints refers to the angle formed between the forearm and the upper arm of the arm.

[0042] In a specific implementation, the angle of the left knee joint and the angle of the right knee joint reflect whether the legs are in a flexed state and the degree of flexion. The angle of the left and right knee joints refers to the angle formed between the calf and the thigh of the leg.

[0043] In a specific implementation, the position, distance and angle formed by the above-mentioned skeletal nodes can be used to accurately judge the posture of the tester. If the vertical jump meets the specifications, it is determined to be valid, and the vertical jump can be judged very quickly and accurately.

[0044] In an embodiment of the present invention, determining that the test person's in-situ vertical jump is valid also includes: judging whether the test person's vertical jump height meets the jumping specifications through the positions of the skeleton points in the test person's jumping action in the characteristic action image; judging whether the test person's knee flexion action meets the landing specifications through the positions of the skeleton points in the test person's landing action in the characteristic action image; the landing specifications include: the angle of the left knee joint and the angle of the right knee joint, both of which are less than the third angle threshold; if the take-off specifications, the jumping specifications and the landing specifications are met at the same time, then the test person's in-situ vertical jump is determined to be valid.

[0045] In a specific implementation, judging whether the vertical jump height of the tester meets the jumping specification may be judging whether the distance between the left toe or the right toe and the ground meets the height requirement.

[0046] In a specific implementation, during the landing action, the abdominal contraction action can also be detected. During the landing action, the angle formed by the central lumbar point and the lines connecting the shoulders and hips is between 0 and 180 degrees.

[0047] In a specific implementation, when analyzing the take-off specifications of the characteristic action images of the test person's vertical jump action, it can be required that the human body postures in all the characteristic action images meet the take-off specifications, or it can be required that the human body postures in some (for example, one quarter to three quarters) of the characteristic action images meet the take-off specifications, or it can be required that the human body postures in one characteristic action image meet the take-off specifications. The specific settings can be made according to actual conditions.

[0048] In the embodiment of the present invention, as described in step 4, the jumping action evaluation unit determines whether the torso swing amplitude of the test person meets the vertical jump specification through the positions of the skeleton points in the test person's jumping action in the characteristic action image.

[0049] In an embodiment of the present invention, the vertical jump standard (used to judge the stability of vertical jump) includes: when judging the left and right swing of the trunk, the x-axis distance from the left shoulder to the vertical line where the central lumbar point is located (a straight line perpendicular to the ground, the same below), and the x-axis distance from the right shoulder to the vertical line where the central lumbar point is located, and the x-axis distance deviation is less than the third distance threshold; when judging the forward and backward swing of the trunk, the y-axis distance from the left shoulder to the vertical line where the central lumbar point is located, and the y-axis distance from the right shoulder to the vertical line where the central lumbar point is located, and the y-axis distance deviation is less than the fourth distance threshold; when judging the vertical axis swing of the trunk, the vertical swing angle formed by the throat and the crotch and the central lumbar point, respectively, and the vertical swing angle change is less than the fourth angle threshold; with the central lumbar point as the coordinate origin, the direction of the tester's shoulders as the x-axis, the direction from the heel to the toes as the y-axis, and the height direction as the z-axis, a three-dimensional coordinate system is established.

[0050] In the specific implementation, as described in step 4, the jumping stability of the tester should be judged based on the left-right / front-back / vertical swing angles of the centers of the two shoulder joints to determine whether the squatting torso swings left-right, whether the squatting torso swings front-back, whether the squatting vertical axis swings, whether the airborne torso swings left-right, whether the airborne torso swings front-back, whether the airborne vertical axis swings, whether the cushioning torso swings left-right, whether the cushioning torso swings front-back, and whether the cushioning vertical axis swings.

[0051] In a specific implementation, the left and right swing of the trunk is judged by the above-mentioned x-axis distance deviation, that is, |X a –X e | and |X b –X e |The deviation between.

[0052] In a specific implementation, the forward and backward swing of the trunk is judged by the above-mentioned y-axis distance deviation, that is, |Y a –Y e | and |Y b –Y e The judgment can also be made by determining that the y-axis distance between the left hip joint point and the vertical line where the central lumbar point is located, and the y-axis distance between the right hip joint point and the vertical line where the central lumbar point is located, and the deviation of the two y-axis distances is less than the fourth distance threshold.

[0053] In a specific implementation, the vertical axis swing of the trunk is judged by the change of the vertical swing angle, which is formed by the line connecting the throat and the crotch, and the line connecting the throat and the central lumbar point. It can also be determined by detecting the angle formed by the central lumbar point and the left (right) shoulder point and the left (right) hip joint point respectively by the left body camera, and whether the change of the angle exceeds the corresponding threshold.

[0054] In an embodiment of the present invention, as described in step 5, the compensation evaluation unit determines whether the rotation amplitude of the test person's hip, knee and foot meets the no-motion compensation specification through the positions of the skeletal points in the test person's take-off and landing actions in the characteristic motion image, thereby determining whether the test person's hip, knee and foot have motion compensation.

[0055] In the embodiment of the present invention, the no-motion compensation specification includes: a first hip-knee rotation angle formed by a line connecting the left hip joint, the left knee joint and the central lumbar point, a second hip-knee rotation angle formed by a line connecting the right hip joint, the right knee joint and the central lumbar point, a change in the first hip-knee rotation angle and the second hip-knee rotation angle is less than a fifth angle threshold; a first y-axis distance between the left heel and the left toe, a second y-axis distance between the right heel and the right toe, and a change in the first y-axis distance and the second y-axis distance is less than a fifth distance threshold.

[0056] In a specific implementation, the change in the first hip-knee rotation angle is less than the fifth angle threshold, and the change in the second hip-knee rotation angle is less than the fifth angle threshold, which reflects whether the hip and knee joints rotate and the rotation range during take-off and landing.

[0057] In a specific implementation, the change in the first y-axis distance is less than the fifth distance threshold, and the change in the second y-axis distance is less than the fifth distance threshold, which reflects whether the foot rotates and the rotation amplitude during take-off and landing.

[0058] In specific implementation, movement compensation reflects whether the test person uses the wrong muscle parts and joints of the body to exert and receive force when performing vertical jump movements, indicating whether the test person may cause serious damage to the leg muscles and ankles due to excess movement compensation.

[0059] In a specific implementation, the above distance difference (deviation) and angle change can be obtained by comparing the positions of human body bones in multiple characteristic action images.

[0060] In a specific implementation, the first distance threshold, the second distance threshold, the swing distance threshold, the third distance threshold, the fourth distance threshold, the fifth distance threshold, the first angle threshold, the second angle threshold, the third angle threshold, the fourth angle threshold and the fifth angle threshold, etc., can be set according to the actual application scenario, which reflects the strictness of the vertical jump specification.

[0061] In the embodiment of the present invention, as described in step 6, the comprehensive evaluation unit provides the judgment result of the tester's action evaluation.

[0062] In the embodiment of the present invention, under the premise that the vertical jump in place of the test person is valid, if both the vertical jump standard and the no-motion compensation standard are met, the test person's action is judged to be a standard vertical jump in place action.

[0063] In the embodiment of the present invention, the judgment result described in step 6 mainly includes the following scoring points: the height score corresponding to the vertical jump height is calculated; the stability score corresponding to the trunk swing amplitude is calculated; and the non-motion compensation score corresponding to the rotation amplitude of the hip, knee and foot is calculated. The height score, stability score and non-motion compensation score are weighted and added to calculate the quality score of the vertical jump in place.

[0064] In specific implementation, the present invention is scored according to the importance and quality of the action during the vertical jump in situ, and is divided into three major items and 12 minor items, wherein the three major items include jumping height, jumping stability and no movement compensation, the jumping height accounts for n1 / N of the total score, and the jumping stability evaluation includes the left and right swing of the squatting trunk, the front and back swing of the squatting, the vertical axis swing of the squatting, the left and right swing of the trunk in the air, the front and back swing of the jump, the vertical axis swing in the air, the left and right swing of the buffering trunk, the front and back swing of the buffering, and the vertical axis swing of the buffering, each accounting for n2 / 9N of the total score, totaling n2 / N. The non-movement compensation action includes the rotation of the hip and knee joints during take-off and landing, and the rotation of the foot during take-off and landing, each accounting for n3 / 2N of the total score, totaling n3 / N, and the total score is the jumping height score * 5 / 12 + the jumping stability item score * 1 / 3 + the movement compensation item * 1 / 4.

[0065] The present invention also provides an in-situ vertical jump action evaluation device, comprising: an image acquisition unit, a skeleton point position determination unit, a take-off action evaluation unit, a jumping action evaluation unit, a motion compensation evaluation unit and a comprehensive evaluation unit, wherein: the image acquisition unit is used to test a person performing an in-situ vertical jump action and obtain a sequence of motion images of the person performing the action; the skeleton point position determination unit is used to select a feature motion image from the motion image sequence and determine the position of the skeleton point of the person performing the action in the motion image; the take-off action evaluation unit is used to determine the position of the skeleton point of the person performing the take-off action in the feature motion image. Whether the posture and position of the tester's arms and legs meet the take-off specifications. If so, the tester's vertical jump in place is deemed valid; the jumping action evaluation unit is used to determine whether the tester's torso swing amplitude meets the vertical jump specifications through the positions of the skeletal points in the tester's jumping action in the characteristic action image; the motion compensation evaluation unit is used to determine whether the rotation amplitude of the tester's hip, knee and foot meets the no-motion compensation specifications through the positions of the skeletal points in the tester's take-off and landing actions in the characteristic action image; the comprehensive evaluation unit is used to provide a judgment result of the tester's action evaluation.

[0066] In the embodiments of the present invention, the method, step or function performed by the unit of the in-situ vertical jump action evaluation device provided by the present invention can refer to the method, step or function in the in-situ vertical jump action evaluation method provided by the present invention, and technicians in the relevant fields can make various changes according to the purpose of this patent without changing its core purpose.

Claims

1. A method for evaluating vertical jump in situ, It is characterized in that include: The tester performs a vertical jump in place, and obtains a sequence of action images of the tester during the action; Selecting a characteristic action image from the action image sequence, and determining the position of the skeleton point of the test person in the action image; Through the position of the skeleton points in the tester's take-off action in the characteristic action image, it is judged whether the posture and position of the tester's arms and legs meet the take-off specifications. If they meet the take-off specifications, the tester's vertical jump in place is deemed valid; By using the positions of the skeleton points in the jumping action of the test person in the characteristic action image, it is determined whether the torso swing amplitude of the test person meets the vertical jump specification; By using the positions of the skeletal points of the tester's take-off and landing movements in the characteristic action images, it is determined whether the rotation range of the tester's hip, knee and foot meets the no-motion compensation standard; Provides the judgment results of the tester's action evaluation.

2. The method for evaluating vertical jump in situ according to claim 1, It is characterized in that The bone point positions include: central lumbar point, left shoulder, right shoulder, left heel, right heel, left toe, right toe, left elbow joint, right elbow joint, left knee joint, right knee joint, left hip joint, right hip joint, throat, and crotch.

3. The method for evaluating vertical jump in situ according to claim 2, It is characterized in that The step of selecting a characteristic action image from the action image sequence comprises: Select the action image in which the test person is at the highest height from the ground from the action image sequence, take the shooting time of the action image with the highest height from the ground as the first reference time, and select the jumping action images before and after the first reference time at a predetermined time interval; select the action image in which the test person squats the lowest from the action image sequence, take the shooting time of the action image in which the test person squats the lowest as the second reference time, and select the taking-off action images before and after the second reference time at a predetermined time interval; select the action image in which the test person lands and touches the ground from the action image sequence, take the shooting time of the action image in which the test person lands and touches the ground as the third reference time, and select the landing action image after the third reference time at a predetermined time interval; and use the selected jumping action image, taking-off action image and landing action image as feature action images.

4. The method for evaluating vertical jump in situ according to claim 2, It is characterized in that The step of judging whether the posture and position of the tester's arms and legs meet the take-off specification includes: The take-off specifications include: the distance from the left heel to the central lumbar point and the distance from the right heel to the central lumbar point, the distance deviation is less than a first distance threshold; the distance from the left heel to the right heel is greater than the distance from the left shoulder to the right shoulder, and the distance difference is less than a second distance threshold; the distance from the left shoulder to the left elbow joint and the distance from the right shoulder to the right elbow joint are both greater than the swing distance threshold; the angle of the left elbow joint and the angle of the right elbow joint are both less than the first angle threshold; the angle of the left knee joint and the angle of the right knee joint are both less than the second angle threshold.

5. The method for evaluating vertical jump in situ according to claim 2, It is characterized in that If the take-off specifications are met, the tester's vertical jump in situ is deemed valid, which also includes: By judging whether the vertical jump height of the test person meets the jumping specifications through the positions of the skeletal points in the test person's jumping action in the characteristic action image; judging whether the knee flexion action of the test person meets the landing specifications through the positions of the skeletal points in the test person's landing action in the characteristic action image; the landing specifications include: the angle of the left knee joint and the angle of the right knee joint, both of which are less than the third angle threshold; if the take-off specifications, jumping specifications and landing specifications are met at the same time, the test person's vertical jump in place is deemed valid.

6. The method for evaluating vertical jump in situ according to claim 2, It is characterized in that The step of judging whether the trunk swing amplitude of the tester meets the vertical jump specification includes: The vertical jump specifications include: the x-axis distance from the left shoulder to the vertical line where the central lumbar point is located, and the x-axis distance from the right shoulder to the vertical line where the central lumbar point is located, and the x-axis distance is less than the third distance threshold; the y-axis distance from the left shoulder to the vertical line where the central lumbar point is located, and the y-axis distance from the right shoulder to the vertical line where the central lumbar point is located, and the y-axis distance is less than the fourth distance threshold; the vertical swing angle formed by the throat and the crotch and the central lumbar point respectively, and the vertical swing angle change is less than the fourth angle threshold; with the central lumbar point as the coordinate origin, the direction of the tester's shoulders as the x-axis, the direction from the heel to the toes as the y-axis, and the height direction as the z-axis, a three-dimensional coordinate system is established.

7. The method for evaluating vertical jump in situ according to claim 6, It is characterized in that The determination of whether the rotation range of the tester's hip, knee and foot meets the no-motion compensation standard includes: The no-motion compensation norms include: a first hip-knee rotation angle formed by a line connecting the left hip joint, the left knee joint and the central lumbar point, a second hip-knee rotation angle formed by a line connecting the right hip joint, the right knee joint and the central lumbar point, a change between the first hip-knee rotation angle and the second hip-knee rotation angle is less than a fifth angle threshold; a first y-axis distance between the left heel and the left toe, a second y-axis distance between the right heel and the right toe, and a change between the first y-axis distance and the second y-axis distance is less than a fifth distance threshold.

8. The method for evaluating vertical jump in situ according to claim 2, It is characterized in that The determination result of the action evaluation provided to the tester includes: On the premise that the tester's vertical jump in place is valid, if it meets both the vertical jump standard and the no-motion compensation standard, the tester's action is judged to be a standard vertical jump in place action.

9. The method for evaluating vertical jump in situ according to claim 8, It is characterized in that The step of providing the judgment result of the tester's action evaluation comprises: The corresponding height score is calculated according to the vertical jump height, the corresponding stability score is calculated according to the trunk swing amplitude, and the corresponding non-motion compensation score is calculated according to the rotation amplitude of the hip, knee and foot. The height score, stability score and non-motion compensation score are weighted and added together to calculate the quality score of the in-situ vertical jump action.

10. A device for evaluating vertical jump in situ, It is characterized in that include: An image acquisition unit, a skeleton point position determination unit, a take-off action evaluation unit, a jumping action evaluation unit, a motion compensation evaluation unit and a comprehensive evaluation unit, wherein: The image acquisition unit is used to acquire a sequence of action images of the tester during the tester's action when the tester performs an in-situ vertical jump action; The skeleton point position determination unit is used to select a characteristic action image from the action image sequence and determine the skeleton point position of the test person in the action image; The take-off action evaluation unit is used to determine whether the posture and position of the tester's arms and legs meet the take-off specification based on the position of the skeleton points in the tester's take-off action in the characteristic action image, and if they meet the take-off specification, the tester's vertical jump in place is deemed valid; The jumping action evaluation unit is used to determine whether the test person's trunk swing amplitude meets the vertical jump specification based on the position of the skeleton points in the test person's jumping action in the characteristic action image; The motion compensation evaluation unit is used to determine whether the rotation range of the test person's hip, knee and foot meets the no-motion compensation specification based on the positions of the skeletal points in the test person's take-off and landing movements in the characteristic motion image; The comprehensive evaluation unit is used to provide the judgment result of the tester's action evaluation.

Citation Information

Patent Citations

  • Evaluation method for sports postures of teenagers

    CN114159760A

  • Limb movement function evaluation method and device

    CN115153518A

  • Rehabilitation exercise quality evaluation method and system

    CN115841864A

  • Jumping landing error evaluation method and device

    CN117037278A

  • Motion evaluation method and device, heterogeneous AI chip, equipment and storage medium

    CN117065310A