Track and field training information acquisition and feedback system based on digital runway

Abstract

The invention discloses a digital runway-based track and field training information acquisition and feedback system, which comprises hardware equipment and a software system, and is characterized in that the hardware equipment comprises a digital runway main body, a high-speed camera main body and an intelligent bracelet main body; the software system comprises a data acquisition module, a data transmission module, a data processing module, a data analysis module, a real-time feedback module and a training report generation module, the digital runway basic sensor, the high-speed camera and the intelligent bracelet are combined for data acquisition, and the pressure sensor, the speed sensor and the stride sensor can acquire athlete sports mechanics data; the high-speed camera can capture motion posture details, the intelligent bracelet monitors the physiological state, multi-source data fusion enables collected information to be more comprehensive, and limitation of a single data source is avoided; through comprehensive data acquisition and accurate feedback, the athletes can optimize technical actions, the coaches can reasonably arrange training content and intensity, and the athletic performance of the athletes is effectively improved.

Description

Technical Field

[0001] The invention relates to a digital runway system device, in particular to a track and field training information collection and feedback system based on a digital runway, belonging to the technical field of sports training. Background Art

[0002] During track and field training, coaches and athletes need to develop corresponding plans based on their usual training to improve their levels. Digital tracks came into being. Digital tracks mainly achieve their functions by distributing large-area flexible pressure array sensors, force platforms and other equipment under the plastic track and connecting them to computers.

[0003] After searching, Chinese patent number CN1818569A discloses a track and field training information collection and feedback system based on digital runway, which collects information of athletes during exercise and transmits this information to a computer through a field bus. After kinematic and dynamic analysis, the training expert system will provide targeted training suggestions and feedback to athletes in real time through a wireless network. However, the above patented product can only analyze according to the shape, time, ground pushing force and support force, and air time of the athlete's sole contacting the runway. When measuring the athlete's action posture, there is a lack of professional equipment and technology, and most of them rely on the naked eye observation of the coach. Key posture information such as the angle of the stride leg when hurdling and the position of the body's center of gravity at the moment of take-off in the triple jump cannot be accurately obtained by the naked eye alone, resulting in the inability to conduct in-depth analysis and scientific guidance of the athlete's technical movements. The coach cannot formulate a highly personalized scientific training plan based on sufficient data. The athlete also lacks timely and accurate feedback to adjust his own state and technical movements during the training process, resulting in poor training results and difficult breakthroughs in sports performance. Summary of the invention

[0004] The purpose of the present invention is to provide a track and field training information collection and feedback system based on a digital track in order to solve the above problems.

[0005] The present invention achieves the above-mentioned purpose through the following technical solutions: a track and field training information collection and feedback system based on a digital runway, including hardware equipment and a software system, wherein the hardware equipment includes a digital runway body, a high-speed camera body and a smart bracelet body, and the software system includes a data collection module, a data transmission module, a data processing module, a data analysis module, a real-time feedback module and a training report generation module; The digital track body adopts a multi-layer composite structure, and the smart bracelet body is worn on the athlete's wrist; The data acquisition module includes a digital runway basic sensor submodule, a high-speed camera submodule and a smart bracelet submodule; the data processing module includes a data cleaning submodule, a video processing submodule and a data integration submodule; the data analysis module includes a sprint analysis submodule, a hurdle analysis submodule and a triple jump analysis submodule; the real-time feedback module includes a sprint feedback submodule, a hurdle feedback submodule and a triple jump feedback submodule; the training report generation module includes a data statistics submodule, a technical analysis submodule and a training suggestion generation submodule.

[0006] Preferably, the digital track body comprises, from the upper layer to the lower layer, a sports surface layer, a sensor integration layer and a buffer support layer, and the sensor integration layer comprises a pressure sensor body and a speed sensor body.

[0007] Preferably, the high-speed camera body is arranged at the starting point, the end point, the curve and each key action node position of the digital runway body.

[0008] Preferably, the digital runway basic sensor submodule includes a pressure sensor submodule, a speed sensor submodule, and a stride sensor submodule. The pressure sensor converts the pressure signal into an electrical signal using the piezoelectric effect. The speed sensor measures the speed based on the laser Doppler effect or the electromagnetic induction principle. The stride sensor calculates the stride by measuring the time interval and speed of the steps.

[0009] Preferably, the high-speed camera submodule includes an image acquisition submodule, an image recognition submodule, and a motion tracking submodule. The image acquisition submodule uses photoelectric conversion technology to acquire images. The image recognition submodule uses a target detection algorithm based on a convolutional neural network, such as FasterR-CNN and YOLO series algorithms, to identify the body parts and movements of athletes. The motion tracking submodule uses Kalman filtering, particle filtering and other algorithms to track the identified targets in real time.

[0010] Preferably, the smart bracelet sub-module includes an acceleration sensor sub-module, a heart rate sensor sub-module, a blood oxygen sensor sub-module, and a positioning sub-module. The acceleration sensor uses micro-electromechanical system technology to sense acceleration, the heart rate sensor uses photoelectric volumetric pulse wave technology to measure heart rate, the blood oxygen sensor monitors blood oxygen saturation based on PPG technology, and the positioning sub-module uses positioning algorithms such as the global positioning system and the Beidou satellite navigation system to determine the motion trajectory.

[0011] Preferably, the data cleaning submodule uses a statistical outlier detection algorithm, such as the 3σ criterion, to identify and remove erroneous data caused by sensor failure, interference, etc. The video processing submodule includes a key frame extraction submodule and a feature labeling submodule. The key frame extraction uses an algorithm based on lens change detection, and the feature labeling adopts a combination of manual labeling and machine learning assisted labeling. The machine learning assisted labeling can use a transfer learning algorithm and perform fine-tuning based on an existing action labeling data set. The data integration submodule uses a data fusion algorithm, such as weighted average fusion and Kalman filter fusion.

[0012] Preferably, the sprint analysis submodule uses a deep learning algorithm, such as a long short-term memory network to analyze the time series relationship between the speed change curve and the arm swing and leg lifting movements, and uses a principal component analysis algorithm to reduce the dimension of the sprint data to find out the main factors affecting the speed. The hurdle analysis submodule is based on a hurdle technology analysis submodule of a dynamic model, uses the principles of Newtonian mechanics to establish the force and motion equations when hurdling, and uses a support vector machine algorithm to classify data such as the take-off leg angle and the swing leg movement to judge the quality of the hurdle technology. The triple jump analysis submodule uses a genetic algorithm to optimize the take-off rhythm and force distribution pattern, finds the optimal solution by simulating the biological evolution process, and uses a cluster analysis algorithm, such as the K-Means algorithm, to cluster the jump data of different athletes to find out similar take-off patterns.

[0013] Preferably, the sprint feedback submodule, the hurdle feedback submodule and the triple jump feedback submodule all use a simple threshold judgment algorithm to trigger a feedback reminder when the monitoring data exceeds or falls below a set threshold.

[0014] Preferably, the data statistics submodule uses basic statistical functions and formulas to calculate average values, maximum values, minimum values, etc., the technical analysis submodule uses professional sports mechanics knowledge and existing analysis models, and the training suggestion generation submodule uses a rule-based reasoning algorithm to generate specific suggestions based on the gap between the athlete's data and the goal and combined with training experience rules.

[0015] The present invention has the following beneficial effects: 1. Combine digital track basic sensors, high-speed cameras and smart bracelets for data collection. Pressure, speed and stride sensors can obtain athletes' kinematic data, high-speed cameras can capture details of movements and postures, and smart bracelets monitor physiological status. Multi-source data fusion makes the collected information more comprehensive and avoids the limitations of a single data source; 2. Use advanced physical sensing principles and algorithms, such as the piezoelectric effect of pressure sensors, the laser Doppler effect or electromagnetic induction principle of speed sensors, and the CNN algorithm of image recognition, to ensure the accuracy of data measurement and analysis, and provide a reliable basis for subsequent training and analysis; 3. The real-time feedback module can promptly remind athletes of physiological abnormalities and technical deviations through smart bracelets. Athletes can make immediate adjustments during training to avoid repeating wrong movements, reduce the risk of injury, and improve training efficiency; 4. The detailed reports and personalized suggestions provided by the training report generation module can help coaches formulate scientific training plans based on the specific conditions of athletes, and athletes can also clearly understand their own shortcomings and conduct targeted training; 5. Through comprehensive data collection and accurate feedback, athletes can optimize their technical movements, such as sprint arm swing, hurdle take-off leg angle, triple jump landing angle, etc.; coaches can reasonably arrange training content and intensity, such as starting reaction training, special strength training, etc., thereby effectively improving athletes' athletic performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A stereogram showing the overall structure of the hardware equipment of a track and field training information collection and feedback system based on a digital runway proposed by the present invention; Figure 2 A stereoscopic diagram of the main structure of a digital runway of a track and field training information collection and feedback system based on a digital runway proposed by the present invention; Figure 3 The present invention provides a flow chart of a track and field training information collection and feedback system based on a digital track.

[0017] In the figure: 1. digital runway body; 101. sports surface layer; 102. sensor integration layer; 103. buffer support layer; 104. pressure sensor body; 105. speed sensor body; 2. high-speed camera body; 3. smart bracelet body. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] Embodiment 1: Reference Figure 1-3 , a track and field training information collection and feedback system based on a digital runway, including hardware equipment and a software system, characterized in that: the hardware equipment includes a digital runway body 1, a high-speed camera body 2 and a smart bracelet body 3, and the software system includes a data collection module, a data transmission module, a data processing module, a data analysis module, a real-time feedback module and a training report generation module; The digital track body 1 adopts a multi-layer composite structure, and the smart bracelet body 3 is worn on the athlete's wrist.

[0020] The digital track body 1 includes a sports surface layer 101 , a sensor integration layer 102 and a buffer support layer 103 from the upper layer to the lower layer. The sensor integration layer 102 includes a pressure sensor body 104 and a speed sensor body 105 .

[0021] The high-speed camera body 2 is arranged at the starting point, the end point, the curve and each key action node position of the digital runway body 1.

[0022] In this embodiment, it should be noted that the sports surface layer 101 is made of sports-specific materials that are highly wear-resistant, non-slip and have good elasticity, such as specially formulated rubber or high-performance polyurethane, to ensure the safety and comfort of athletes and assist in exerting force.

[0023] The sensor integration layer 102 serves as the core layer and integrates a variety of high-precision sensors. The pressure sensor measures the distribution and time of the ground-pushing force, the speed sensor captures the speed change, and the stride sensor monitors the stride size, providing a basis for motion data analysis.

[0024] The buffer support layer 103 uses a high-elasticity and high-strength buffer material, such as high-strength rubber or special shock-absorbing foam, to provide stable support for the runway, absorb impact force, and protect athlete joints and middle-layer sensors.

[0025] The pressure sensor body 104, speed sensor body 105 and stride sensor in the sensor integration layer 102 collect relevant data of athletes in the process of starting, running, jumping, etc. in real time. The pressure sensor body 104 accurately senses the ground pushing force, the speed sensor body 105 uses the laser Doppler effect or electromagnetic induction technology to capture the speed change, and the stride sensor monitors the stride in real time. For example, in sprint training, the pressure sensor can measure the peak value of the athlete's double-foot ground pushing force at the moment of starting; in hurdle training, it can record the ground pushing force at the moment of crossing the hurdle; in triple jump training, the ground pushing force data of each jump can be obtained.

[0026] Multiple high-speed camera bodies 2 are arranged at the starting point, end point, bends and key action nodes of the runway to form a camera array. The high-speed camera body 2 captures the athlete's motion picture at a high frame rate, tracks the athlete's body parts through image recognition and motion tracking technology, and calculates parameters such as stride length, stride frequency, and action angle. In short-distance running, the start, acceleration, and sprinting actions are captured; when hurdling, the start, hurdles, and sprinting are recorded; and in triple jump, the details of each stage of action such as run-up, take-off, take-off, and landing are recorded throughout the process.

[0027] The smart bracelet body 3 worn by the athlete integrates an acceleration sensor, a heart rate sensor, a blood oxygen sensor, etc. The acceleration sensor monitors the change of motion acceleration, the heart rate sensor measures the heart rate in real time, and the blood oxygen sensor monitors the blood oxygen saturation. At the same time, the smart bracelet records the motion trajectory through the built-in positioning module. During training, the athlete's physiological data is monitored in real time, such as the heart rate and blood oxygen changes before and after each jump in the sprint stage, the hurdle process, and the triple jump.

[0028] Embodiment 2: Different from the first embodiment, referring to 1-3, this embodiment also has the following further contents: the data acquisition module includes a digital runway basic sensor submodule, a high-speed camera submodule and a smart bracelet submodule, the data processing module includes a data cleaning submodule, a video processing submodule and a data integration submodule, the data analysis module includes a sprint analysis submodule, a hurdle analysis submodule and a triple jump analysis submodule, the real-time feedback module includes a sprint feedback submodule, a hurdle feedback submodule and a triple jump feedback submodule, and the training report generation module includes a data statistics submodule, a technical analysis submodule and a training suggestion generation submodule.

[0029] The digital runway basic sensor submodule includes a pressure sensor submodule, a speed sensor submodule, and a stride sensor submodule. The pressure sensor uses the piezoelectric effect to convert the pressure signal into an electrical signal. The speed sensor measures the speed based on the laser Doppler effect or the electromagnetic induction principle. The stride sensor calculates the stride by measuring the time interval and speed of the steps.

[0030] The high-speed camera submodule includes an image acquisition submodule, an image recognition submodule, and a motion tracking submodule. The image acquisition submodule uses photoelectric conversion technology to acquire images. The image recognition submodule uses a target detection algorithm based on a convolutional neural network, such as FasterR-CNN and the YOLO series of algorithms, to identify the body parts and movements of athletes. The motion tracking submodule uses Kalman filtering, particle filtering and other algorithms to track the identified targets in real time.

[0031] The smart bracelet sub-modules include an accelerometer sub-module, a heart rate sensor sub-module, a blood oxygen sensor sub-module, and a positioning sub-module. The accelerometer uses micro-electromechanical system technology to sense acceleration, the heart rate sensor uses photoelectric volumetric pulse wave technology to measure heart rate, the blood oxygen sensor monitors blood oxygen saturation based on PPG technology, and the positioning sub-module uses positioning algorithms such as the Global Positioning System and the Beidou Satellite Navigation System to determine the motion trajectory.

[0032] The data cleaning submodule uses statistical outlier detection algorithms, such as the 3σ criterion, to identify and remove erroneous data caused by sensor failure, interference, etc. The video processing submodule includes a key frame extraction submodule and a feature labeling submodule. Key frame extraction uses an algorithm based on lens change detection. Feature labeling uses a combination of manual labeling and machine learning-assisted labeling. Machine learning-assisted labeling can use a transfer learning algorithm and perform fine-tuning based on existing action labeling data sets. The data integration submodule uses data fusion algorithms, such as weighted average fusion and Kalman filter fusion.

[0033] The sprint analysis submodule uses deep learning algorithms, such as long short-term memory network to analyze the time series relationship between the speed change curve and the arm swing and leg lifting movements, and uses the principal component analysis algorithm to reduce the dimension of the sprint data to find out the main factors affecting the speed. The hurdle analysis submodule is based on the hurdle technology analysis submodule of the dynamic model. It uses the principles of Newtonian mechanics to establish the force and motion equations when crossing the hurdle, and uses the support vector machine algorithm to classify data such as the take-off leg angle and the swing leg movement to judge the quality of the hurdle technology. The triple jump analysis submodule uses a genetic algorithm to optimize the take-off rhythm and force distribution pattern, and finds the optimal solution by simulating the biological evolution process. It uses cluster analysis algorithms, such as the K-Means algorithm, to cluster the jump data of different athletes to find similar take-off patterns.

[0034] The sprint feedback submodule, hurdle feedback submodule and triple jump feedback submodule all use simple threshold judgment algorithms. When the monitored data exceeds or falls below the set threshold, a feedback reminder is triggered. The data statistics submodule uses basic statistical functions and formulas to calculate the average, maximum, minimum, etc. The technical analysis submodule uses professional sports mechanics knowledge and existing analysis models. The training suggestion generation submodule uses a rule-based reasoning algorithm to generate specific suggestions based on the gap between the athlete's data and the goal and combined with training experience rules.

[0035] In this embodiment, it should be noted that the basic sensor data of the digital runway is transmitted to the data processing module via wired (high-speed Ethernet) or wireless (5G, etc.). The video data captured by the high-speed camera is initially encoded and transmitted via a high-speed network. The data collected by the smart bracelet is transmitted to a nearby relay device via Bluetooth, and then uploaded to the data processing module by the relay device via Wi-Fi or 5G network.

[0036] After receiving the data, the data processing module first cleans the data to remove erroneous data and noise data caused by sensor failure, interference, etc. It extracts key frames and annotates features of the video data, and organizes all types of data into a unified format to prepare for subsequent data analysis.

[0037] The real-time feedback module uses the smart bracelet body 3 to remind athletes of physiological abnormalities and technical movement deviations in real time. When the heart rate is detected to be too high and the blood oxygen saturation is lower than the normal range, the smart bracelet body 3 vibrates and sounds prompts to remind athletes to adjust their exercise intensity. During the training process, if it is analyzed that the athlete's technical movements do not meet the standards, such as insufficient arm swing in the sprint stage, abnormal leg angle when crossing the hurdles, and deviation in the landing angle of the triple jump, the smart bracelet body 3 will vibrate and display prompt information to guide the athlete to adjust the movement in time.

[0038] After training, the training report generation module generates a detailed training report, including sports data statistics, such as speed, distance, number of steps, stride, strength, etc.; technical analysis, such as posture assessment, force pattern analysis; training suggestions, based on the athlete's training data and goals, provide personalized training plan adjustment suggestions, such as increasing or decreasing training intensity, optimizing training project combinations, improving technical movements, etc. Coaches and athletes can view training reports through computers or mobile applications to better plan subsequent training.

[0039] Embodiment three: The digital track system is used in sprint training: Data collection: In the 100-meter sprint training, the basic sensors of the digital track body 1 are fully operational. At the start moment, the pressure sensor measures the peak force of the athlete's feet pushing the ground to 4000N, and the speed sensor captures the start acceleration of 5m / s² within 0.1 seconds. In the acceleration stage, the stride sensor records the average stride gradually increasing from 2.0 meters to 2.2 meters, and the speed sensor shows that the speed continues to rise, reaching a maximum of 10.5m / s. The high-speed camera body 2 accurately captures the entire process of the athlete's start, acceleration, and sprint at a frame rate of 300 frames per second. The smart bracelet body 3 monitors the athlete's physiological data in real time. The heart rate in the preparation stage is 70 beats / minute, which instantly soars to 130 beats / minute at the start. The heart rate in the sprint stage is maintained at 180-190 beats / minute, and the blood oxygen saturation fluctuates between 94% and 96%.

[0040] Data transmission and processing: The basic sensor data of the digital runway body 1 is quickly transmitted to the data processing module with the help of the 5G network. The high-definition video data shot by the high-speed camera body 2 is initially encoded and compressed, and then stably transmitted through the high-speed 5G network. The data collected by the smart bracelet body 3 is first transmitted to a nearby relay device via Bluetooth, and then uploaded to the data processing module via Wi-Fi. The data processing module strictly cleans all types of data to remove abnormal values ​​caused by factors such as signal interference, such as occasional jump data from the speed sensor. Key frames are extracted from the video data, and key stages such as starting, acceleration, mid-run, and sprinting are accurately marked.

[0041] Data analysis and feedback: The data analysis module uses advanced deep learning algorithms to deeply analyze data. Through comprehensive analysis of the video and sensor data of the high-speed camera body 2, it is determined that the athlete's arm swing amplitude is 8° less than the standard value during the sprint stage, which affects the speed improvement. The real-time feedback module immediately vibrates and displays prompt information through the smart bracelet body 3 to remind the athlete to increase the arm swing amplitude. After the training, the training report generation module generates a detailed report. The report shows that the athlete's starting reaction time is 0.16 seconds, which is 0.02 seconds slower than the average reaction time of excellent athletes; the average stride is 2.1 meters, which is different from the ideal stride of 2.2 meters; the maximum speed is 10.5m / s, and there is still room for improvement from the target speed of 11m / s. Based on the report, the coach arranges 150 sets of starting reaction training every day, with the goal of shortening the reaction time to less than 0.14 seconds; 4 special arm swing exercises per week, 200 sets each time, focusing on correcting the arm swing amplitude.

[0042] Embodiment 4: The digital track system is used in hurdle training: Data collection: In the 110-meter hurdles training, the basic sensors of the digital track body 1 record key data in real time. At the start, the pressure sensor measured the peak pressure of the athlete's right foot pushing the ground to 3800N, and the speed sensor captured the start acceleration of 4.8m / s² in 0.13 seconds. When crossing the hurdles, the stride sensor recorded an average stride of 3.45 meters, and the speed sensor showed that the speed was stable at 8.0m / s at the moment of crossing the hurdle. The high-speed camera body 2 records the whole process of the athlete's start, acceleration, hurdles, sprints, etc. at a frame rate of 250 frames per second. The smart bracelet body 3 monitors the athlete's physiological data in real time. The heart rate in the preparation stage is 75 times / minute, which instantly increases to 128 times / minute at the start. The heart rate is maintained at 165-175 times / minute during the hurdles, and the blood oxygen saturation fluctuates between 95% and 97%.

[0043] Data transmission and processing: The data of each sensor is collected and sent to the data processing module according to the established transmission method. The data processing module calibrates the data of pressure sensors, speed sensors, etc. to ensure the accuracy of the data. The feature extraction of the video of the high-speed camera body 2 is carried out to accurately identify key information such as the angle of the starting leg, the posture of crossing the hurdle, and the action of going down the hurdle.

[0044] Data analysis and feedback: The data analysis module deeply analyzes the data by building a professional sports model and finds that the angle of the athlete's starting leg is 5° less than the standard value, which leads to unstable center of gravity when crossing the hurdle and affects the speed of crossing the hurdle. During the training process, once the real-time feedback module detects that the stride fluctuation exceeds ±0.2 meters, it will immediately remind the athlete to adjust the pace rhythm through the smart bracelet body 3. After the training, the training report shows that the athlete's starting reaction time is 0.17 seconds, and the average hurdle speed is 8.0m / s, which is different from the target speed of 8.5m / s; the rhythm deviation is mainly reflected between the 4th and 5th hurdles, which takes 0.12 seconds more than the optimal rhythm. According to the report, the coach arranges 120 sets of starting reaction training every day, with the goal of shortening the reaction time to less than 0.15 seconds; three times a week, special exercises for the starting leg, 180 sets each time, focusing on correcting the angle of the starting leg.

[0045] Embodiment five: The digital track system is used in triple jump training: Data collection: In triple jump training, the basic sensor of the digital runway body 1 records an average stride of 2.6 meters in the first 5 steps during the run-up phase, with the speed accelerating from 0 to 7.0 m / s. The stride of the last 5 steps is stable at 2.8 meters, and the speed increases to 8.5 m / s. In the first jump (single-leg hop), the horizontal speed at the moment of take-off is 8.2 m / s, the vertical speed is 2.5 m / s, and the take-off angle is 18°; the horizontal speed of the second jump (stride jump) is maintained at 7.8 m / s, the vertical speed is 2.2 m / s, and the take-off angle is 17°; the horizontal speed of the third jump (jump) is 7.5 m / s, the vertical speed is 2.8 m / s, and the take-off angle is 20°. The high-speed camera body 2 records the details of the movements in each stage, including the run-up, take-off, vacant, and landing. The smart bracelet body 3 detected that the athlete's heart rate was 100 beats / minute before the jump, rose to 145 beats / minute after the first jump, and reached 170 beats / minute after the third jump, and the blood oxygen saturation gradually dropped from 96% to 93%.

[0046] Data transmission and processing: Data is transmitted to the data processing module according to the process, and the module comprehensively integrates and pre-processes the data. The video shot by the high-speed camera body 2 is analyzed frame by frame to accurately extract key feature points such as the center of gravity position of the body when taking off, the body posture when flying, and the buffering action when landing.

[0047] Data analysis and feedback: The data analysis module uses big data analysis technology to compare the data of excellent athletes and finds that the landing angle of the athlete's second jump is 4° larger than the standard value, resulting in a significant loss of horizontal speed. During the training process, when the real-time feedback module detects that the fluctuation of the running-up amplitude exceeds ±0.15 meters, the smart bracelet body 3 reminds the athlete to adjust the rhythm. After the training, the training report shows that the average running-up speed reaches 8.5m / s, the first jump distance is 10.0 meters, the second jump is 10.5 meters, and the third jump is 11.2 meters. According to the report, the coach arranges 130 second jump landing angle adjustment exercises every day, setting a goal for each exercise to correct the landing angle to the standard value; and increases leg strength training 3 times a week to improve take-off strength.

Claims

1. A track and field training information collection and feedback system based on a digital runway, comprising hardware equipment and a software system, characterized in that: The hardware device comprises a digital runway body (1), a high-speed camera body (2) and a smart bracelet body (3), and the software system comprises a data acquisition module, a data transmission module, a data processing module, a data analysis module, a real-time feedback module and a training report generation module; The digital running track body (1) adopts a multi-layer composite structure, and the smart bracelet body (3) is worn on the athlete's wrist; The data acquisition module includes a digital runway basic sensor submodule, a high-speed camera submodule and a smart bracelet submodule; the data processing module includes a data cleaning submodule, a video processing submodule and a data integration submodule; the data analysis module includes a sprint analysis submodule, a hurdle analysis submodule and a triple jump analysis submodule; the real-time feedback module includes a sprint feedback submodule, a hurdle feedback submodule and a triple jump feedback submodule; the training report generation module includes a data statistics submodule, a technical analysis submodule and a training suggestion generation submodule.

2. The track and field training information collection and feedback system based on a digital runway according to claim 1, characterized in that: The digital track body (1) comprises, from the upper layer to the lower layer, a sports surface layer (101), a sensor integration layer (102) and a buffer support layer (103), wherein the sensor integration layer (102) comprises a pressure sensor body (104) and a speed sensor body (105).

3. The track and field training information collection and feedback system based on a digital runway according to claim 1, characterized in that: The high-speed camera body (2) is arranged at the starting point, the end point, the bend and each key action node position of the digital runway body (1).

4. The track and field training information collection and feedback system based on a digital runway according to claim 1, characterized in that: The digital runway basic sensor submodule includes a pressure sensor submodule, a speed sensor submodule, and a stride sensor submodule. The pressure sensor converts the pressure signal into an electrical signal using the piezoelectric effect. The speed sensor measures the speed based on the laser Doppler effect or the electromagnetic induction principle. The stride sensor calculates the stride by measuring the time interval and speed of the steps.

5. A track and field training information collection and feedback system based on a digital runway according to claim 4, characterized in that: The high-speed camera submodule includes an image acquisition submodule, an image recognition submodule, and a motion tracking submodule. The image acquisition submodule uses photoelectric conversion technology to acquire images. The image recognition submodule uses a target detection algorithm based on a convolutional neural network, such as FasterR-CNN and YOLO series algorithms, to identify the body parts and movements of athletes. The motion tracking submodule uses Kalman filtering, particle filtering and other algorithms to track the identified targets in real time.

6. A track and field training information collection and feedback system based on a digital runway according to claim 5, characterized in that: The smart bracelet submodule includes an acceleration sensor submodule, a heart rate sensor submodule, a blood oxygen sensor submodule, and a positioning submodule. The acceleration sensor uses micro-electromechanical system technology to sense acceleration, the heart rate sensor uses photoelectric volumetric pulse wave technology to measure heart rate, the blood oxygen sensor monitors blood oxygen saturation based on PPG technology, and the positioning submodule uses positioning algorithms such as the global positioning system and the Beidou satellite navigation system to determine the motion trajectory.

7. A track and field training information collection and feedback system based on a digital runway according to claim 6, characterized in that: The data cleaning submodule uses a statistical outlier detection algorithm, such as the 3σ criterion, to identify and remove erroneous data caused by sensor failure, interference, etc. The video processing submodule includes a key frame extraction submodule and a feature labeling submodule. The key frame extraction uses an algorithm based on lens change detection. Feature labeling uses a combination of manual labeling and machine learning assisted labeling. Machine learning assisted labeling can use a transfer learning algorithm and perform fine-tuning based on an existing action labeling data set. The data integration submodule uses a data fusion algorithm, such as weighted average fusion and Kalman filter fusion.

8. The track and field training information collection and feedback system based on a digital runway according to claim 7, characterized in that: The sprint analysis submodule uses deep learning algorithms, such as long short-term memory network to analyze the time series relationship between the speed change curve and the arm swing and leg lifting movements, and uses the principal component analysis algorithm to reduce the dimension of the sprint data to find out the main factors affecting the speed. The hurdle analysis submodule is based on the hurdle technology analysis submodule of the dynamic model, uses the principles of Newtonian mechanics to establish the force and motion equations when crossing the hurdle, and uses the support vector machine algorithm to classify data such as the take-off leg angle and the swing leg movement to judge the quality of the hurdle technology. The triple jump analysis submodule uses a genetic algorithm to optimize the take-off rhythm and force distribution pattern, finds the optimal solution by simulating the biological evolution process, and uses a cluster analysis algorithm, such as the K-Means algorithm, to cluster the jump data of different athletes to find similar take-off patterns.

9. A track and field training information collection and feedback system based on a digital runway according to claim 8, characterized in that: The sprint feedback submodule, the hurdle feedback submodule and the triple jump feedback submodule all use a simple threshold judgment algorithm to trigger a feedback reminder when the monitoring data exceeds or falls below a set threshold.

10. A track and field training information collection and feedback system based on a digital track according to claim 9, characterized in that: The data statistics submodule uses basic statistical functions and formulas to calculate average values, maximum values, minimum values, etc. The technical analysis submodule uses professional sports mechanics knowledge and existing analysis models. The training suggestion generation submodule uses a rule-based reasoning algorithm to generate specific suggestions based on the gap between the athlete's data and the goal and combined with training experience rules.

Images