Student comprehensive physical interesting exercise test method based on motion quality grading and adaptive combination
By using a comprehensive physical fitness test method for students based on motion quality grading and adaptive combination, and by dynamically adjusting test items using facial recognition and motion key point monitoring, the method solves the problems of fun and adaptability of existing physical fitness test methods, and achieves a more realistic and comprehensive assessment of students' physical fitness.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGDONG PROPHET BIG DATA CO LTD
- Filing Date
- 2026-01-19
- Publication Date
- 2026-04-24
AI Technical Summary
Existing student physical fitness testing methods lack fun, have rigid difficulty settings, fail to adapt to individual differences, lack monitoring of movement quality, and affect children's enthusiasm for participation and the authenticity of test results.
Student information is obtained through facial recognition, and the initial difficulty and incremental score are calculated based on the national physical fitness and health standards. The coordinates of key points of the movements are monitored in real time, and the test items are dynamically adjusted by an adaptive combination strategy, which integrates fun design and adaptive difficulty adjustment.
It improves children's participation and attention, dynamically adjusts test content, evaluates the quality of movements in real time, provides personalized physical fitness test plans, and enhances the authenticity and comprehensiveness of test results.
Smart Images

Figure CN121920676A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of physical fitness testing technology, specifically to a method for testing students' comprehensive physical fitness through fun sports activities based on movement quality grading and adaptive combination. Background Technology
[0002] With the increasing emphasis placed on students' physical health by education authorities, physical fitness testing has become an important means of assessing the physical fitness of teenagers. Traditional student physical fitness tests mainly rely on the National Student Physical Fitness and Health Standards, using standardized items such as rope skipping, running, sit-and-reach, and sit-ups to assess students' physical fitness levels. However, with the development of technology and the diversification of testing needs, existing physical fitness testing methods have revealed many shortcomings in practical applications.
[0003] In recent years, physical fitness testing methods based on computer vision and deep learning technologies have gradually emerged. Chinese patent CN116650922A discloses a comprehensive physical fitness testing method for teenagers based on deep learning. This method detects human body position and estimates posture, combining temporal image information captured by a mobile phone camera and the coordinates of key posture points to test and evaluate seven qualities: strength, flexibility, balance, agility, coordination, endurance, and speed. Chinese patent CN119831804A proposes a comprehensive physical fitness testing auxiliary system for sports, which acquires students' historical physical fitness test data and inputs it into a physical fitness test evaluation model to achieve student physical fitness test training management. Chinese patent CN107730414A discloses a wireless sports teaching system that can record students' sports test scores in real time and convert them into scores and grades according to national standards. In addition, Chinese patent CN114642424A discloses a physical fitness assessment method based on motion-sensing interaction technology, which uses a 3D motion imaging sensor to collect user activity information for physical fitness assessment. Chinese patent CN105608467B proposes a contactless student physical fitness assessment method based on Kinect, which uses skeletal nodes to record changes in the position of human joints to detect the standardization of movements.
[0004] Despite some progress in the intelligence and automation of physical fitness testing, existing technologies still suffer from the following shortcomings: First, the testing methods are monotonous and boring. Traditional repetitive movements such as rope skipping and running lack interest, leading to low participation and poor concentration among children, severely affecting the authenticity and validity of the test results. Second, the difficulty levels are too rigid. Existing methods generally use uniform testing standards, failing to effectively adapt to individual differences among children. High-achieving students find the tests too easy and lack challenge, while lower-achieving students are prone to frustration. Finally, there is a lack of effective process monitoring mechanisms. Existing technologies mainly focus on recording the final test scores and cannot capture and analyze key process information such as movement quality and completion rate in real time, failing to provide accurate data support for personalized training. These problems prevent existing physical fitness testing methods from fully playing their role in assessing and improving students' physical health. Summary of the Invention
[0005] To address the technical problems of existing children's physical fitness testing methods, such as monotonous testing formats, fixed difficulty settings, and lack of process monitoring, and to achieve the technical effects of increasing children's participation, enabling adaptive adjustment of difficulty, and establishing a real-time movement quality monitoring mechanism, this invention provides a student comprehensive physical fitness fun sports test method based on movement quality grading and adaptive combination.
[0006] The objective of this invention is achieved through the following technical solution: a method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination, comprising the following steps: S1. Pre-store students' personal information and basic physical fitness data in the testing system. When the test starts, acquire students' facial images through an image acquisition device and perform facial recognition. Based on the recognition results, retrieve the student's scores in four national student physical fitness and health standard tests: one-minute rope skipping, 50-meter sprint, sit-and-reach, and one-minute sit-ups, respectively. s, g r, g p, g u; S2. Based on the results of the National Student Physical Fitness and Health Standard Test, the initial difficulty scores of the five test items, squatting, boxing, jumping, high knee raising, and side bending, are calculated using a threshold grading algorithm and are recorded as gp1, gp2, gp3, gp4, and gp5, respectively. The difficulty score of each item is determined by the corresponding National Student Physical Fitness and Health Standard Test item score according to the preset grading threshold. S3. Based on the national student physical fitness and health standard test scores, calculate the initial incremental scores for the five test items: squatting, boxing, jumping, high knees, and side bending, and denoted as ga1, ga2, ga3, ga4, and ga5, respectively. The incremental scores are used for project scheduling in the subsequent adaptive phase. The calculation formula is based on the weighted sum of the national test scores and the allocation of the constant ns1. S4, the first 5N sets are trained in a fixed order, with each 5 sets constituting a cycle. The exercises performed are squats, boxing, jumping, high knees, and side bends. During each set, the system visually captures the student's key movement coordinates: neck coordinates (nx). i,j ny i.y ), left elbow coordinates (elx) i,j ,ely i.j Right elbow coordinates (erx) i,j ery i.j ), left wrist coordinates (alx) i,j aly i.j ), right wrist coordinates (arx) i,j ary i.j ), left hip coordinates (ulx) i,j uly i.j ), right hip coordinates (urx) i,j ,ury i.j Left knee coordinates (klx) i,j ,kly i.j ), right knee coordinates (krx) i,j ,kry i.j ), left foot coordinates (flx) i,j fly i.j ), right foot coordinate (frx) i,j fry i.j Based on the key point coordinate data, the quality score gf for each action is calculated. i,j ; Calculate the movement score g for each training set. i Combined score of action quality (gz) i ; S5. For the five test items, accumulate the quality scores of the first N groups respectively, and calculate the quality scores of each item such as squatting, boxing, jumping, high knee, and side bending as the average of the comprehensive quality scores of the corresponding items in the first N groups, which are GB1, GB2, GB3, GB4, and GB5 respectively, as the basis for the selection of items in the subsequent adaptive phase. S6. Perform the first adaptive selection. During the training of the 5N+1th group, based on the cumulative quality scores gb1, gb2, gb3, gb4, and gb5 of the five items, use the minimum score selection strategy to determine the training item for this session, selecting the item with the lowest score as the training item, and setting the remaining incremental score ga of the corresponding item. i Subtract 1; S7. After each training session, the quality score of the item is updated in real time. Based on the optimization principle of minimizing the item selection score and the remaining incremental score, the next training item is dynamically selected. When i is greater than 5N+1, the selection score of the previous training item is updated in real time. The training item is selected based on the principle of minimizing the item selection score divided by the remaining incremental score plus 1. The remaining incremental score of the selected item is decremented by 1, while the remaining incremental scores of other items remain unchanged. Specifically, gb1 i =(gb1 i-1 +gz i ) / 2 when the (i-1)th group item is squatting, otherwise gb1 i =gb1 i-1 The scores for other items will be updated according to the same rules; S8. When the total number of sets i = 5N + ns1 is completed, the training process ends, and the total action score g for all sets is accumulated. i When the 5N+ns1 group of training is completed, the test is stopped, the total comprehensive physical fitness score G is calculated, the student's physical fitness level is evaluated based on the total comprehensive physical fitness score G, and the student's comprehensive physical fitness status is judged according to the level of the total comprehensive physical fitness score G.
[0007] Preferably, in step S2 , Excellent is indicated by "4"; good is indicated by "3" for scores of 80-90; passing is indicated by "2" for scores of 60-80; and failing is indicated by "1" for scores below 60.
[0008] Preferably, in step S3 , Where ns1 is a positive integer, representing the maximum increment constant.
[0009] Preferably, in step S4, the standard score for the completion of the j-th movement in the i-th group of follow-up training is calculated, with the item being squat gf1. i,j The project is boxing gf2 i,j The project is Jump GF3 i,j The project is high knees gf4 i,j The project involves side bending over (gf5). i,j: in, , gf11 i,j Score the quality of the squatting motion. , gf21 i,j Score the quality of boxing movements , gf31i,j Score the quality of the jump action , c1 is a set first constant, representing the pixel ordinate of the jump line within the scene. It is a fixed constant for each scene.
[0010] gf41 i,j Score the quality of the high knee movement. , gf51 i,j Score the quality of the side bend movement. , ts1 is the set first judgment threshold, ts2 is the set second judgment threshold, ts3 is the set third judgment threshold, ts4 is the set fourth judgment threshold, ts5 is the set fifth judgment threshold, ts6 is the set sixth judgment threshold, i=1,...,5N + ns1; TS1 is calculated by having students who scored 90 or higher on rope skipping or running physical tests complete squatting movements. The system calculates the student's squatting movement quality score using the method described above, taking one-third of the minimum score distribution. TS2 is based on students who score 90 or higher on a sit-up test and perform boxing movements. The system calculates the student's boxing movement quality score using the method described above, taking one-sixth of the minimum score distribution. TS3 is calculated by having students with a jump rope test score of 90 or higher complete the jump action. The system calculates the student's jump action quality score based on the above method, taking one-third of the minimum score distribution. TS4 is calculated by having students who score 90 or higher on sit-ups or running physical tests complete high knee exercises. The system calculates the student's high knee exercise quality score based on the above method, taking one-sixth of the minimum score distribution. TS5 is calculated by having students with a sit-and-reach test score of 90 or higher complete the side bend movement. The system calculates the quality score of the student's side bend movement based on the above method, taking one-third of the minimum score distribution. gf11 i,jFirst, determine whether a squatting motion is valid by checking the relative positions of the neck and knees on the vertical axis (the neck should be higher than the knees to prevent the person from lying on the ground), the relative positions of the feet and hips (the hips should be on top of the feet to prevent the person from sitting instead of squatting), and the relative positions of the knees and hips (the hips should be lower than the knees to ensure that it is a squatting motion and not just a bending over). After determining that a squatting motion is valid, determine the degree of squatting based on the ratio of the distance from the hips to the knees to the length of the lower leg on the vertical axis to obtain the squatting quality. gf21 i,j First, determine whether the person's movements meet the standard for punching. This is done by judging the relative positional relationship between the neck and hip bones and the elbow and wrist. When the person's movements are punching, the quality of the boxing action is judged by judging the ratio of the wrist displacement to the elbow displacement on the horizontal axis. gf31 i,j First, determine whether the person is jumping with straight legs by judging the relative positions of the feet, knees, hips, and neck. If the person is jumping with straight legs, judge the quality of the jump by calculating the ratio of the jump height to the person's height. gf41 i,j First, determine whether the person meets the standard for raising their leg. This is done by judging the relative position of the foot and the other knee. When the person is raising their leg, the quality of the high leg raise is judged by the ratio of the height of the two feet to the length of the supporting leg. gf51 i,j First, determine whether the person meets the standard for bending over. This is done by judging the relative position of the neck and hip bones. When the person is bending over, the quality of the side bend is judged by the ratio of the distance the neck moves laterally to the width of the hips.
[0011] Preferably, in step S4, after calculating the standard score for the j-th movement in the i-th group of follow-up practice, the movement score g in the i-th group of follow-up practice is then calculated. i: , in, .
[0012] Preferably, in step S4, the motion score g in the i-th group of follow-up practice is calculated. i Then, calculate the overall score g of the movement quality in the i-th group of follow-up exercises. i: , in, , Where, n i This represents the number of actions completed in the i-th group of practice sessions. The ratio of the sum of the quality of each group of movements to the minimum quality score of outstanding students is used as the overall quality score. Dividing by the minimum quality score of outstanding students is a process similar to "normalization" to ensure that the overall quality score of each group of movements can be quantitatively compared within the same range.
[0013] Preferably, in step S5, the quality scores for each item—squatting, boxing, jumping, high knees, and side bends—are calculated as the average of the comprehensive quality scores for the first N groups of corresponding items, respectively: GB1, GB2, GB3, GB4, GB... 5: .
[0014] Preferably, in step S6, when performing the i=5N+1th test group, the scores for the selection of items such as squatting, boxing, jumping, high knees, and side bends are calculated as gb1 respectively. i, gb2 i, gb3 i, gb4 i, gb5 i; gb1 i =gb1 gb2 i =gb2 gb3 i =gb3 gb4 i =gb4 gb5 i =gb5 The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i, ga2 i, ga3 i, ga4 i, ga5 i : When GB1 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is squatting, ga1 i =max(0,ga1-1) ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB2 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is boxing, ga1 i =ga1 ga2 i =max(0,ga2-1) ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB3 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is jumping, ga1 i =ga1 ga2 i =ga2 ga3 i =max(0,ga3-1) ga4 i =ga4 ga5 i =ga5 Conversely, when GB4 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is high knee raises, ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =max(0,ga4-1) ga5 i =ga5 Conversely, when GB5 i =min(gb1) i, gb2 i, gb3 i,gb4 i, gb5 i When the i-th test item is a side bend, ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =max(0,ga5-1).
[0015] Preferably, in step S7, when i>5N+1, the selection scores for squatting, boxing, jumping, high knees, and side bends are calculated as gb1 respectively. i, gb2 i, gb3 i, gb4 i, gb5 i: , The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i, ga2 i, ga3 i, ga4 i, ga5 i , when At that time, the i-th test item was squatting. ga1 i =max(0,ga1) i -1 - 1) ga2 i =ga2 i -1 ga3 i =ga3 i -1 ga4 i =ga4 i -1 ga5 i =ga5 i -1 Conversely At that time, the test item for group i was boxing. ga1 i =ga1 i -1 ga2 i =max(0,ga2) i -1 - 1) ga3 i =ga3 i -1 ga4 i =ga4 i -1 ga5 i =ga5 i -1 Conversely When the i-th test item is jumping, ga1 i =ga1 i -1 ga2 i =ga2 i -1 ga3 i =max(0,ga3) i -1 - 1) ga4 i =ga4 i -1 ga5 i =ga5 i -1 Conversely At that time, the test item for the i-th group was high knee raises. ga1 i =ga1 i -1 ga2 i =ga2 i -1 ga3 i =ga3 i -1 ga4 i =max(0,ga4) i -1 - 1) ga5 i =ga5 i -1 Conversely At that time, the test item for the i-th group was side bending over. ga1 i =ga1 i -1 ga2 i =ga2 i -1 ga3 i =ga3 i -1 ga4 i =ga4 i -1 ga5 i =max(0,ga5) i -1 - 1).
[0016] Preferably, in step S8, when the training for group i = 5N + ns1 is completed, the test is stopped, and the student's total physical fitness score G is calculated: , The student's overall physical fitness level is judged based on their total score (G).
[0017] The beneficial effects of this invention are as follows: During testing, the system acquires student information through facial recognition and automatically matches initial action plans based on students' national physical fitness test scores, effectively improving children's participation enthusiasm and concentration, and enhancing the authenticity of test results; By calculating the initial difficulty score and initial incremental score of five test items based on the national student physical fitness and health standard test scores, adaptive difficulty adjustment is achieved, which can dynamically adjust the test content according to individual differences among children, avoiding problems such as high-ability students feeling bored and low-ability students experiencing frustration due to uniform standards; By capturing the coordinates of key parts of the student's body in real time through visual detection, and calculating the completion standard score and quality score of five actions based on the key point coordinates, a real-time action quality monitoring mechanism is established, which can capture key information in the process of action completion, not only recording the final score but also evaluating the quality of the action; By adopting an adaptive testing process, the first 5N groups are the basic test groups, and subsequent test items are dynamically selected based on the selected project score and the remaining exercise score, integrating a complete testing system of fun design, adaptive difficulty adjustment, comprehensive evaluation, and intelligent feedback, providing a new technical solution for children's physical fitness testing. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating the testing method of the present invention. Detailed Implementation
[0019] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.
[0020] See Figure 1 A method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combinations includes the following steps: S1. Pre-store students' personal information and basic physical fitness data in the testing system. When the test starts, acquire students' facial images through an image acquisition device and perform facial recognition. Based on the recognition results, retrieve the student's scores in four national student physical fitness and health standard tests: one-minute rope skipping, 50-meter sprint, sit-and-reach, and one-minute sit-ups, respectively. s, g r, g p, g u; S2. Based on the National Student Physical Fitness and Health Standard Test scores, a threshold grading algorithm is used to calculate the initial difficulty scores for five test items: squatting, boxing, jumping, high knees, and side bends, which are denoted as gp1, gp2, gp3, gp4, and gp5, respectively. gp1 is a grading calculation based on the average of rope skipping and running scores, gp2 is a grading calculation based on sit-up scores, gp3 is a grading calculation based on rope skipping scores, gp4 is a grading calculation based on running scores, and gp5 is a grading calculation based on sit-and-reach scores. The difficulty score for each item is determined by the corresponding National Student Physical Fitness and Health Standard Test item scores according to a preset grading threshold. S3. Based on the National Student Physical Fitness and Health Standard Test scores, calculate the initial incremental scores for the five test items: squatting, boxing, jumping, high knees, and side bends, denoted as ga1, ga2, ga3, ga4, and ga5 respectively. ga2 is a weighted calculation based on sit-up scores, ga3 is a weighted calculation based on rope skipping scores, ga4 is a weighted calculation based on running and sit-up scores, and ga5 is a weighted calculation based on sit-and-reach scores. ga1 is... ga1 = ns1 minus the difference in the incremental scores of the other four items ns1-ga2-ga3-ga4-ga5; Incremental scores are used for project scheduling in the subsequent adaptive phase. The calculation formula is based on the weighted sum of national test scores and the allocation of the constant ns1. S4, the first 5N sets are trained in a fixed order, with each 5 sets constituting a cycle. The exercises performed are squats, boxing, jumping, high knees, and side bends. During each set, the system visually captures the student's key movement coordinates: neck coordinates (nx). i,j ny i.y ), left elbow coordinates (elx) i,j ,ely i.j Right elbow coordinates (erx) i,j ery i.j ), left wrist coordinates (alx) i,j aly i.j ), right wrist coordinates (arx) i,j ary i.j ), left hip coordinates (ulx) i,j uly i.j ), right hip coordinates (urx) i,j ,ury i.j Left knee coordinates (klx) i,j ,kly i.j ), right knee coordinates (krx) i,j ,kry i.j ), left foot coordinates (flx) i,j fly i.j ), right foot coordinate (frx) i,j fryi.j Based on the key point coordinate data, the quality score gf for each action is calculated. i,j ; Calculate the movement score g for each training set. i Combined score of action quality (gz) i ; S5. For the five test items, accumulate the quality scores of the first N groups respectively, calculate the cumulative quality score of each item, and calculate the quality scores of each item such as squatting, boxing, jumping, high knee, and side bending as the average of the comprehensive quality scores of the corresponding items in the first N groups, which are GB1, GB2, GB3, GB4, and GB5 respectively, as the basis for the selection of items in the subsequent adaptive phase. S6. Perform the first adaptive selection. During the training of the 5N+1th group, based on the cumulative quality scores gb1, gb2, gb3, gb4, and gb5 of the five items, use the minimum score selection strategy to determine the training item for this session, selecting the item with the lowest score as the training item, and setting the remaining incremental score ga of the corresponding item. i Subtract 1; S7. After each training session, the quality score of the item is updated in real time. Based on the optimization principle of minimizing the item selection score and the remaining incremental score, the next training item is dynamically selected. When i is greater than 5N+1, the selection score of the previous training item is updated in real time. The training item is selected based on the principle of minimizing the item selection score divided by the remaining incremental score plus 1. The remaining incremental score of the selected item is decremented by 1, while the remaining incremental scores of other items remain unchanged. Specifically, gb1 i =(gb1 i-1 +gz i ) / 2 when the (i-1)th group item is squatting, otherwise gb1 i =gb1 i-1 The scores for other items will be updated according to the same rules; S8. When the total number of sets i = 5N + ns1 is completed, the training process ends, and the total action score g for all sets is accumulated. i When the 5N+ns1 group of training is completed, the test is stopped, the total comprehensive physical fitness score G is calculated, and the student's physical fitness level is evaluated based on the total comprehensive physical fitness score G. The student's comprehensive physical fitness status is judged based on the score.
[0021] In step S2, , In this evaluation, “4” represents “Excellent”; “3” represents “Good” for scores of 80-90; “2” represents “Pass” for scores of 60-80; and “1” represents “Fail” for scores less than 60.
[0022] In step S3, , Where ns1 is a positive integer, representing the maximum increment constant.
[0023] Preferably, in step S4, the standard score for the completion of the j-th movement in the i-th group of follow-up training is calculated, with the item being squat gf1. i,j The project is boxing gf2 i,j The project is Jump GF3 i,j The project is high knees gf4 i,j The project involves side bending over (gf5). i,j : in, , gf11 i,j The quality score for the squatting motion is calculated based on the positional relationship between the hip, knee, and foot. , gf21 i,j The score for the quality of a boxing move is calculated based on the positional relationship between the wrist, elbow, and hip bones. , gf31 i,j The score for the quality of a jump is calculated based on the ratio of takeoff height to height. , c1 is a set first constant, representing the pixel ordinate of the jump line within the scene, and is a fixed constant for each scene. , gf41 i,j The quality score for the high knee movement is calculated based on the ratio of the height difference between the two feet to the length of the supporting leg. , gf51 i,j The quality score for the side-bending movement is calculated based on the ratio of the lateral movement distance of the neck to the width of the hips. , ts1 is the set first judgment threshold, ts2 is the set second judgment threshold, ts3 is the set third judgment threshold, ts4 is the set fourth judgment threshold, ts5 is the set fifth judgment threshold, ts6 is the set sixth judgment threshold, i=1,...,5N+ns1; TS1 is calculated by having students who scored 90 or higher on rope skipping or running physical tests complete squatting movements. The system calculates the quality score of the students' squatting movements by taking one-third of the minimum score distribution. TS2 is calculated by having students who scored 90 or higher on a sit-up test perform boxing movements. The score for the quality of the students' boxing movements is calculated by taking one-sixth of the minimum value of the score distribution. TS3 is calculated by having students with a jump rope test score of 90 or higher complete the jump action, and the system calculates the student's jump action quality score by taking one-third of the minimum score distribution. TS4 is calculated by having students who score 90 or higher on sit-ups or running tests complete high knee exercises. The system calculates the student's high knee exercise quality score by taking one-sixth of the minimum score distribution. TS5 is calculated by having students who scored 90 or higher on the sit-and-reach test complete the side bend movement. The system calculates the quality score of the student's side bend movement by taking one-third of the minimum score distribution. gf11 i,j First, determine whether a squatting motion is valid by checking the relative positions of the neck and knees on the vertical axis (the neck should be higher than the knees to prevent the person from lying on the ground), the relative positions of the feet and hips (the hips should be on top of the feet to prevent the person from sitting instead of squatting), and the relative positions of the knees and hips (the hips should be lower than the knees to ensure that it is a squatting motion and not just bending over). After determining that a squatting motion is valid, determine the degree of squatting based on the ratio of the distance from the hips to the knees to the length of the lower leg on the vertical axis to obtain the squatting quality. gf21 i,j First, determine whether the person's movements meet the standard for punching. This is done by judging the relative positional relationship between the neck and hip bones and the elbow and wrist. When the person's movements are punching, the quality of the boxing action is judged by judging the ratio of the wrist displacement to the elbow displacement on the horizontal axis. gf31 i,j First, determine whether the person is jumping with straight legs by judging the relative positions of the feet, knees, hips, and neck. If the person is jumping with straight legs, judge the quality of the jump by calculating the ratio of the jump height to the person's height. gf41 i,j First, determine whether the person meets the standard for raising their leg. This is done by judging the relative position of the foot and the other knee. When the person is raising their leg, the quality of the high leg raise is judged by the ratio of the height of the two feet to the length of the supporting leg. gf51 i,j First, determine whether the person meets the standard for bending over. This is done by judging the relative position of the neck and hip bones. When the person is bending over, the quality of the side bend is judged by the ratio of the distance the neck moves laterally to the width of the hips.
[0024] In step S4, after calculating the standard score for the j-th movement in the i-th set of follow-up practice, the movement score g in the i-th set of follow-up practice is then calculated. i : , in, , In step S4, the motion score g in the i-th group of follow-up practice is calculated. i Then, calculate the overall score g of the movement quality in the i-th group of follow-up exercises. i : , in, , Where, n i This represents the number of actions completed in the i-th group of practice sessions. The ratio of the sum of the quality of each group of movements to the minimum quality score of outstanding students is used as the overall quality score. Dividing by the minimum quality score of outstanding students is a process similar to "normalization" to ensure that the overall quality score of each group of movements in each project can be quantitatively compared within the same range.
[0025] In step S5, the quality scores for each item—squatting, boxing, jumping, high knees, and side bends—are calculated as the average of the comprehensive quality scores for the first N groups of corresponding items, respectively: GB1, GB2, GB3, GB4, and GB5. .
[0026] In step S6, when performing the i=5N+1th test group, calculate the selection scores for squatting, boxing, jumping, high knees, and side bends, respectively, which are gb1. i GB2 i GB3 i GB4 i GB5 i , gb1 i =gb1 gb2 i =gb2 gb3 i =gb3 gb4 i =gb4 gb5 i =gb5 The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i ga2 i ga3 i ga4 i ga5 i When GB1 i =min(gb1) i, gb2i, gb3 i, gb4 i, gb5 i When the i-th test item is squatting, ga1 i =max(0,ga1-1) ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB2 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is boxing, ga1 i =ga1 ga2 i =max(0,ga2-1) ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB3 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is jumping, ga1 i =ga1 ga2 i =ga2 ga3 i =max(0,ga3-1) ga4 i =ga4 ga5 i =ga5 Conversely, when GB4 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is high knee raises, ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =max(0,ga4-1) ga5 i =ga5 Conversely, when GB5 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i When the i-th test item is a side bend, ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =max(0,ga5-1).
[0027] Preferably, in step S7, when i>5N+1, the selection scores for squatting, boxing, jumping, high knees, and side bends are calculated as gb1 respectively. i, gb2 i, gb3 i, gb4 i, gb5 i :
[0028] The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i, ga2 i, ga3 i, ga4 i, ga5 i : when At that time, the i-th test item was squatting. ga1 i =max(0,ga1) i-1 - 1) ga2 i =ga2 i-1 ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely At that time, the test item for group i was boxing. ga1 i =ga1 i-1 ga2 i =max(0,ga2) i-1 - 1) ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely When the i-th test item is jumping, ga1 i =ga1 i-1 ga2 i =ga2 i-1 ga3 i =max(0,ga3) i-1 - 1) ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely At that time, the test item for the i-th group was high knee raises. ga1 i =ga1 i-1 ga2 i =ga2 i-1 ga3 i =ga3 i-1 ga4 i =max(0,ga4) i-1 - 1) ga5 i =ga5 i-1 Conversely At that time, the test item for the i-th group was side bending over. ga1 i=ga1 i-1 ga2 i =ga2 i-1 ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =max(0,ga5) i-1 - 1) Preferably, in step S8, when the training for group i = 5N + ns1 is completed, the test is stopped, and the student's total physical fitness score G is calculated: , The student's overall physical fitness level is judged based on their total score (G).
[0029] This method ensures the scientific nature and accuracy of the test through a motion quality grading mechanism, and realizes personalized physical fitness test plans through an adaptive combination strategy. It can dynamically adjust the test content according to the student's actual performance, thereby more comprehensively and objectively assessing the student's physical fitness level.
[0030] The above embodiments are preferred implementations of the present invention. In addition, the present invention can be implemented in other ways. Any obvious substitutions without departing from the concept of the present invention are within the protection scope of the present invention.
Claims
1. A method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination, characterized in that: Includes the following steps: S1. Pre-store students' personal information and basic physical fitness data in the testing system. When the test starts, acquire students' facial images through an image acquisition device and perform facial recognition. Based on the recognition results, retrieve the student's scores in four national student physical fitness and health standard tests: one-minute rope skipping, 50-meter sprint, sit-and-reach, and one-minute sit-ups, respectively. s, g r, g p, g u; S2. Based on the results of the National Student Physical Fitness and Health Standard Test, the initial difficulty scores of the five test items, squatting, boxing, jumping, high knee raising, and side bending, are calculated using a threshold grading algorithm and are recorded as gp1, gp2, gp3, gp4, and gp5, respectively. The difficulty score of each item is determined by the corresponding National Student Physical Fitness and Health Standard Test item score according to the preset grading threshold. S3. Based on the national student physical fitness and health standard test scores, calculate the initial incremental scores for the five test items: squatting, boxing, jumping, high knees, and side bending, and denoted as ga1, ga2, ga3, ga4, and ga5, respectively. The incremental scores are used for project scheduling in the subsequent adaptive phase. The calculation formula is based on the weighted sum of the national test scores and the allocation of the constant ns1. S4, the first 5N sets are trained in a fixed order, with each 5 sets constituting a cycle. The exercises performed are squats, boxing, jumping, high knees, and side bends. During each set, the system visually captures the student's key movement coordinates: neck coordinates (nx). i,j ny i.y ), left elbow coordinates (elx) i,j ,ely i.j Right elbow coordinates (erx) i,j ery i.j ), left wrist coordinates (alx) i,j aly i.j ), right wrist coordinates (arx) i,j ary i.j ), left hip coordinates (ulx) i,j uly i.j ), right hip coordinates (urx) i,j ,ury i.j Left knee coordinates (klx) i,j ,kly i.j ), right knee coordinates (krx) i,j ,kry i.j ), left foot coordinates (flx) i,j fly i.j ), right foot coordinate (frx) i,j fry i.j Based on the key point coordinate data, the quality score gf for each action is calculated. i,j ; Calculate the movement score g for each training set. i Combined score of action quality (gz) i ; S5. For the five test items, accumulate the quality scores of the first N groups respectively, and calculate the quality scores of each item such as squatting, boxing, jumping, high knee, and side bending as the average of the comprehensive quality scores of the corresponding items in the first N groups, which are GB1, GB2, GB3, GB4, and GB5 respectively, as the basis for the selection of items in the subsequent adaptive phase. S6. Perform the first adaptive selection. During the training of the 5N+1th group, based on the cumulative quality scores gb1, gb2, gb3, gb4, and gb5 of the five items, use the minimum score selection strategy to determine the training item for this session, selecting the item with the lowest score as the training item, and setting the remaining incremental score ga of the corresponding item. i Subtract 1; S7. After each training session, the quality score of the item is updated in real time. Based on the minimum value optimization principle of "item selection score / (remaining incremental score + 1)", the next training item is dynamically selected. S8. When the total number of sets i = 5N + ns1 is completed, the training process ends, and the total action score g for all sets is accumulated. i Calculate the overall physical fitness score G, and evaluate the student's physical fitness level based on the overall physical fitness score G.
2. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 1, characterized in that: In step S2 , In this evaluation, "4" represents excellent; "3" represents good (80, 90); "2" represents passing (60, 80); and "1" represents failing (less than 60).
3. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 2, characterized in that: In step S3 , Where ns1 is a positive integer, representing the maximum increment constant.
4. The student comprehensive physical fitness fun sports test method based on motion quality grading and adaptive combination according to claim 3, characterized in that: In step S4, calculate the standard score for the j-th movement in the i-th set of follow-up exercises, with the item being squat gf1. i,j, The project is boxing gf2 i,j The project is Jump GF3 i,j, The project is high knees gf4 i,j, The project is a side bend over gf5 i,j: in, , gf11 i,j Score the quality of the squatting motion. , gf21 i,j Score the quality of boxing movements , gf31 i,j Score the quality of the jump action , c1 is a set first constant, representing the pixel ordinate of the jump line within the scene, and is a fixed constant for each scene. , gf41 i,j Score the quality of the high knee movement. , gf51 i,j Score the quality of the side bend movement. , ts1 is the set first judgment threshold, ts2 is the set second judgment threshold, ts3 is the set third judgment threshold, ts4 is the set fourth judgment threshold, ts5 is the set fifth judgment threshold, and ts6 is the set sixth judgment threshold, i = 1, ..., 5N + ns 1。 5. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 4, characterized in that: In step S4, after calculating the standard score for the j-th movement in the i-th set of follow-up practice, the movement score g in the i-th set of follow-up practice is then calculated. i: , in, 。 6. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 5, characterized in that: In step S4, the motion score g in the i-th group of follow-up practice is calculated. i Then, calculate the overall score g of the movement quality in the i-th group of follow-up exercises. i: , in, , Where, n i Let be the number of actions completed in the i-th group of practice.
7. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 6, characterized in that: In step S5, the quality scores for each item—squatting, punching, jumping, high knees, and side bends—are calculated as the average of the comprehensive quality scores for the first N groups of corresponding items, respectively: gb1, gb2, gb3, gb4, gb 5: 。 8. The method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination as described in claim 7, characterized in that: In step S6, when performing the i=5N+1th test group, calculate the selection scores for squatting, boxing, jumping, high knees, and side bends, respectively, which are gb1. i, gb2 i, gb3 i, gb4 i, gb5 i: gb1 i =gb1 gb2 i =gb2 gb3 i =gb3 gb4 i =gb4 gb5 i =gb5 The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i, ga2 i, ga3 i, ga4 i, ga5 i When GB1 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i) At that time, the i-th test item was squatting. ga1 i =max(0,ga1- 1) ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB2 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i) At that time, the test item for group i was boxing. ga1 i =ga1 ga2 i =max(0,ga2- 1) ga3 i =ga3 ga4 i =ga4 ga5 i =ga5 Conversely, when GB3 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i) When the i-th test item is jumping, ga1 i =ga1 ga2 i =ga2 ga3 i =max(0,ga3-1) ga4 i =ga4 ga5 i =ga5 Conversely, when GB4 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i) At that time, the test item for the i-th group was high knee raises. ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =max(0,ga4- 1) ga5 i =ga5 Conversely, when GB5 i =min(gb1) i, gb2 i, gb3 i, gb4 i, gb5 i) At that time, the test item for the i-th group was side bending over. ga1 i =ga1 ga2 i =ga2 ga3 i =ga3 ga4 i =ga4 ga5 i =max(0,ga5- 1)。 9. A method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination, as described in claim 8, is characterized in that: In step S7, when i > 5N+1, calculate the selection scores for squatting, boxing, jumping, high knees, and side bends, which are respectively gb1. i, gb2 i, gb3 i, gb4 i, gb5 i: , The remaining scores for the squat, boxing, jumping, high knee, and side bend events were ga1 and ga1, respectively. i, ga2 i, ga3 i, ga4 i, ga5 i 当 At that time, the i-th test item was squatting. ga1 i =max(0,ga1) i-1 - 1) ga2 i =ga2 i-1 ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely At that time, the test item for group i was boxing. ga1 i =ga1 i-1 ga2 i =max(0,ga2 i-1 - 1) ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely When the i-th test item is jumping, ga1 i =ga1 i-1 ga2 i =ga2 i-1 ga3 i =max(0,ga3) i-1 - 1) ga4 i =ga4 i-1 ga5 i =ga5 i-1 Conversely At that time, the test item for the i-th group was high knee raises. ga1 i =ga1 i-1 ga2 i= ga2 i-1 ga3 i= ga3 i-1 ga4 i= max(0,ga4 i-1 - 1) ga5 i= ga5 i-1 Conversely At that time, the test item for the i-th group was side bending over. ga1 i =ga1 i-1 ga2 i =ga2 i-1 ga3 i =ga3 i-1 ga4 i =ga4 i-1 ga5 i =max(0,ga5 i-1 - 1)。 10. A method for testing students' comprehensive physical fitness through fun sports activities based on motion quality grading and adaptive combination, as described in claim 9, is characterized in that: In step S8, when the training for group i = 5N + ns1 is completed, the test is stopped, and the student's total physical fitness score G is calculated: 。
Citation Information
Patent Citations
Kinect-based contactless student physical fitness assessment method
CN105608467B
Wireless PE teaching system
CN107730414A
Physical fitness assessment method and device based on somatosensory interaction technology
CN114642424A
Deep learning-based teenager physical fitness comprehensive test method and system
CN116650922A
Comprehensive physical fitness test auxiliary system and method for physical exercise
CN119831804A