TW202435093A - Integrated sensing and analyzing system for track and field event hammer throw - Google Patents

Abstract

An integrated sensing and analyzing system, which is for track and field events hammer throw, includes an analyzing device, a drone, a force plate, and a side camera. The analyzing device collects all shoot images and detected data, and establishes a starting time point and an end time point. It also applies a force-exerted position into a first curve chart and a second curve chart. The first curve chart and the second curve chart are respectively obtained from the drone, the side camera, and analysis of the analyzing device, whereby to create visually readable charted information. With the information, problematic points which need to be adjusted can be quickly spotted, whereby precisely adjusting the movement and training for an athlete of the hammer throw event.

Description

Integrated sensor analysis system for track and field ball project

The present invention relates to an integrated sensor analysis system for the track and field chain event, and more particularly to a system that generates easy-to-read graphic information through analysis using an aerial camera, a side camera, and a force plate, thereby accurately adjusting the athlete's limb movements and training methods for chain events.

According to the sports, slingball is a sport that relies heavily on technique and physical strength. The speed of rotation is increased by rotating the body and relying on the weight of the slingball itself. The body must remain stable during the rotation and be adjusted to the best angle before throwing to achieve the farthest throwing effect. Some minor changes in the training process are not easy to detect. In order to make the training results more outstanding, we need to rely on the power of science. However, many sports science equipment are quite expensive, and before use, special clothes or sensors need to be worn to capture the movements of athletes. The slingshot sport can only be performed outdoors when throwing. Sometimes when the weather is hot, if these equipment are worn, the physical consumption may become more intense. If there is a set of affordable equipment that does not require additional sensors, it will not affect the training performance of the athletes, and it will allow coaches to understand the status and problems of athletes, thereby improving the athletes' athletic strength.

In view of this, the inventor has been engaged in the manufacturing, development and design of related products for many years. After careful design and careful evaluation, he finally obtained the present invention which is truly practical.

The technical problem to be solved by the present invention is to provide an integrated sensing and analysis system for track and field ball project in view of the above-mentioned deficiencies in the existing technology.

An analysis device is used to integrate the filming data and the measurement data. The analysis device establishes a common starting time point and an ending time point when starting the measurement. A drone is connected to the analysis device and shoots images above the athlete's head. The analysis device uses an image recognition method to identify the relative position of the athlete's head and the chain ball. The analysis device analyzes the movement trajectory and speed of the chain ball with the athlete's head as the center, and further establishes a first curve chart of the chain ball speed. A force plate is connected to the analysis device and placed in the athlete's foot area to measure the pedaling force. The analysis device determines the axial foot that is continuously subjected to force and the force-applying foot that is instantly subjected to force, and further establishes the athlete's complete A force positioning point for each rotation is formed, and a side camera is connected to the analysis device and set up on the side of the athlete to shoot images. The analysis device identifies the relative positions of the athlete's limb joints using a bone identification method, and further establishes a plurality of second curve graphs based on the analyzed shoulder height, arm elevation angle, hip joint height, knee angle, arm elevation angle, and trunk angle. The analysis device inserts the start time point, end time point, and force positioning point for each rotation into the first curve graph and the second curve graph, so that the problem points that need to be adjusted can be quickly found through the graph, thereby accurately adjusting the athlete's limb movements and training methods for chain ball events.

The analysis device is a set of overlapping graphs of the first curve chart and at least one of the second curve charts, so as to intuitively judge the corresponding position on the second curve chart where the chain ball speed changes significantly, thereby improving the athlete's limb movements in a targeted manner.

The corresponding time axes of the first curve graph and the second curve graph are embedded in the images captured by the drone and the side camera, so that when the analysis device clicks on the curve position on the first curve graph or the second curve graph, the analysis device starts playing the captured images with the corresponding time axis.

The analysis device records the head position of each athlete within the range of the start time point and the end time point, and draws a rotation path line. The rotation path line can be compared with the throwing direction of the chain ball to determine whether the rotation path is deviated.

The analysis device is connected to a grip glove, which is put on the athlete's hand and is used to withstand the centrifugal force of the chain ball. The analysis device analyzes the grip change to determine whether the athlete's hands are stiff and whether the arm is bent. The start time and end time of the analysis device are the start time and end time of the grip glove being subjected to force.

The analysis device is connected to a rear camera, which is mounted behind the athlete to take images. The analysis device uses a backbone identification method to identify the relative positions of the athlete's limb joints, and the images taken by the rear camera compensate for the lack of viewing angle of the side camera, thereby improving the accuracy of the second curve chart.

The analysis device provides input of the chain ball throwing distance each time the athlete performs the chain ball project, and records the chain ball throwing distance in the first curve chart. The chart data of each training is stored in order of date and time, so as to analyze whether the athlete's training method is suitable and understand the degree of improvement.

The analysis device is connected to a cloud database, and the analysis device uploads all the first curve charts and the second curve charts to the cloud database, and the cloud database performs big data analysis, finds abnormal curves based on the big data of the cloud database, and then reminds and marks the abnormal curve parts.

The steps of obtaining the first curve chart by the image recognition method include: a. recording the position of the chain ball in each frame for marking; b. calculating the change of the chain ball position in the previous and next frames to obtain the chain ball displacement; c. dividing the displacement of the speed formula by time to obtain the chain ball speed; d. using the moving average method to eliminate the jagged edges of the original speed curve; e. recording the athlete information and speed data to the analysis device.

The steps of obtaining the second curve chart by the skeleton recognition method include: a. detecting the body position of the athlete in the image frame by using a heat map; b. performing pixel analysis based on the body position to find all limb nodes; c. marking the positions of all limb nodes in each frame; d. calculating the height curve and the angle curve using the corresponding limb nodes; e. eliminating the jagged edges of the height curve or the angle curve by using the moving average method; and f. recording the athlete information, height data and angle data to the analysis device.

The main purpose of the present invention is that the analysis device inserts the starting time point, the ending time point and the force positioning point of each rotation into the first curve graph and the second curve graph, and the first curve graph and the second curve graph are respectively obtained by the aerial camera, the side camera and the analysis device, thereby forming easy-to-read chart information, and quickly finding the problem points that need to be adjusted through the chart method, thereby accurately adjusting the athlete's limb movements and training methods for the chain ball project.

Other objects, advantages and novel features of the present invention will become more apparent from the following detailed description and the accompanying drawings.

In order to enable the review committee to have a deeper understanding and knowledge of the purpose, features and effects of this invention, please cooperate with the following (simple illustration) to describe in detail:

First, please refer to FIG. 1 to FIG. 8 , which show an integrated sensing and analysis system for track and field tee shot events, including: an analysis device 10, an air shot camera 20, a force plate 30, a side shot camera 40, a grip glove 50, a back shot camera 60, and a cloud database 70. The analysis device 10 is used to integrate the shooting data and the measurement data, and analyze the above data and data to produce easy-to-read chart information. The analysis device 10 establishes a common start time point A1 and end time point A5 when starting measurement. The start time point A1 and the end time point A5 can be set by oneself or automatically generated through a detection mechanism. A drone 20 is connected to the analysis device 10 and takes images above the head of the athlete 100. After the images are transmitted to the analysis device 10, the analysis device 10 uses an image recognition method to identify the relative positions of the athlete's 100 head 103 and the chain ball 101, that is, the positions of the athlete's 100 head 103 and the chain ball 101 are marked on each frame of the image. The analysis device 10 analyzes the movement trajectory and speed of the chain ball 101 with the athlete's 100 head 103 as the center, and further establishes a first curve chart S01 of the speed of the chain ball 101 (as shown in FIG. 6). The first curve chart S01 is a curve chart of the speed of the chain ball 101. The curve graph S01 is formed with frames and speed as coordinates, and the start time point A1 and the end time point A5 are marked on the first curve graph S01 as the display range of the speed curve. A force plate 30 is connected to the analysis device 10 and placed under the athlete's 100 foot area for pedaling force measurement, and the force plate 30 will cover the entire pedaling range of the athlete 100, so that each step of the athlete 100 will cause the force plate 30 to generate a pressure signal and transmit it to the analysis device 10. The analysis device 10 determines the axial foot that is continuously stressed and the force-applying foot that is momentarily stressed. This is based on the fact that the limb movements of the ball chain project will cause the force-applying foot to The athlete 100 applies force by stepping on the ground when performing one rotation, and further establishes force positioning points (A2, A3, A4) for completing each rotation of the athlete 100. The force positioning points (A2, A3, A4) are all used to mark the first curve chart S01. A side camera 40 is connected to the analysis device 10 and is set up on the side of the athlete 100 to take images and transmit them to the analysis device 10. The analysis device 10 identifies the relative positions of the limb joints 102 of the athlete 100 by the bone recognition method, and calculates the height L and angle θ between the limb joints 102 from the relative coordinate positions of the limb joints 102. The double angles obtained by analysis are further analyzed. The shoulder height, hip joint height (as shown in FIG. 4 ), knee angle (as shown in FIG. 4 ), arm elevation angle (as shown in FIG. 5 ) and trunk angle are used to establish a plurality of second curve graphs S02 (as shown in FIG. 7 and FIG. 8 ). A grip glove 50 is connected to the analysis device 10. The grip glove 50 is put on the hand of the athlete 100 and is used to withstand the centrifugal force of the chain ball 101. The grip glove 50 transmits the measured grip data to the analysis device 10. The analysis device 10 analyzes the grip change to determine whether the hands of the athlete 100 are stiff and whether the arms are bent. When the hands are too stiff or the arms are bent, the athlete 100 There will be an action of pulling the chain ball 101 back. When the athlete 100 cannot resist the force of the chain ball 101 during the rotation process, it is easy to use the force of the hand to hold the chain ball 101, which will cause the chain ball 101 to fly at a distance that cannot be optimized when the shot is released. This is one of the limb movements that must be improved in training. Furthermore, the starting time point A1 and the ending time point A5 of the analysis device 10 are the starting time point and the ending time point of the force application of the grip glove 50, that is, a method of automatically generating the starting time point A1 and the ending time point A5 through a detection mechanism is provided. A back-up camera 60 is connected to the analysis device 10. The rear camera frame 60 is set behind the athlete 100 to shoot images. The analysis device 10 identifies the relative position of the athlete's 100 limb joints 102 by the skeleton identification method, and the image taken by the rear camera 60 compensates for the lack of the field of view of the side camera 40. When the value of the second curve graph S02 generated by the side camera 40 obviously exceeds the reasonable range, the rear camera 60 replaces part of the curve of the side camera 40. When both the side camera 40 and the rear camera 60 exceed the reasonable range, additional markings are made to reduce misjudgment, thereby improving the accuracy of the second curve graph S02. A cloud database 70 The analysis device 10 is connected to the analysis device 10, and the analysis device 10 uploads all the first curve graph S01 and the second curve graph S02 to the cloud database 70, and the cloud database 70 performs big data analysis, and the cloud database 70 uses big data to determine the abnormal curve, and then a reminder is marked for the abnormal curve part. The abnormal curve refers to the part that needs to be adjusted in the athlete's 100 body movement, thereby reducing the ability requirement for judging the first curve graph S01 and the second curve graph S02, so that the coach or athlete 100 can find the problem points in the body movement of the ball chaining project without a lot of learning.

The steps of obtaining the first curve graph S01 by the image recognition method in this invention include: a. recording the position of the chain ball 101 in each frame for marking; b. calculating the change of the position of the chain ball 101 in the previous and next frames to obtain the displacement of the chain ball 101; c. dividing the displacement of the speed formula by the time to obtain the speed of the chain ball 101; d. using the moving average method to eliminate the jagged edges of the original speed curve; e. recording the athlete 100 information and speed data to the analysis device 10, thereby obtaining the usable first curve graph S01. To further explain, the steps of obtaining the second curve chart S02 by the skeleton recognition method in the present invention include: a. detecting the body position of the athlete 100 in the image frame by using a heat map; b. performing pixel analysis based on the body position to find all limb nodes 102; c. marking the positions of all limb nodes 102 in each frame; d. calculating the height L curve and the angle θ curve based on the corresponding limb nodes 102; e. using the moving average method to eliminate the jagged edges of the height L curve or the angle curve θ; f. recording the athlete 100 information, the height data and the angle data to the analysis device 10, thereby obtaining the athlete 10 0, a plurality of second curve graphs S02 of the double shoulder height curve, the hip joint height curve, the knee angle curve, the arm elevation angle curve and the trunk angle curve, as shown in FIG4, the purpose of calculating the angle θ of the left and right knee bending is to see whether the knee angle θ of the athlete 100 is too large during the rotation process. If the knee angle θ of the athlete is too large during the rotation process, it will affect the throwing stability of the chain ball 101. The purpose of calculating the height L between the left and right feet and the hip joint of the athlete 100 is to see the change of the center of gravity of the athlete 100 during the throwing process. When the height L of the axis foot of the athlete 100 during the exercise is too high, the rotation will be unstable. As shown in FIG5 , the angle θ and height L of the arm swing will affect the acceleration and the angle of the athlete 100. Therefore, by analyzing the angle θ between the arm and the vertical axis and the height L between the shoulder and the wrist, it is possible to understand whether there is a problem during the rotation of the athlete 100 that affects the throwing distance of the chain ball 101. Furthermore, the analysis device 10 inserts the starting time point A1, the ending time point A5, and the force positioning points (A2, A3, A4) of each rotation into the first curve chart S01 and the second curve chart S02, thereby forming easy-to-read chart information, and quickly finding the problem points that need to be adjusted through the chart method, so as to effectively improve the limb movements of the athlete 100 and increase the throwing distance of the chain ball 101, and further 9 and 10, the analyzing device 10 overlays the first curve graph S01 and at least one of the second curve graphs S02, so as to intuitively judge the corresponding position on the second curve graph S02 at the point where the speed of the chain ball 101 changes significantly, thereby improving the body movements of the athlete 100 in a targeted manner, thereby accurately adjusting the body movements and training methods of the athlete 100 in the chain ball event.

As shown in FIG. 3 , the analysis device 10 records the position of the head 103 of each athlete 100 within the range of the starting time point A1 and the ending time point A5, and draws a rotation path line M1. The rotation path line M1 can be compared with the throwing direction of the chain ball 101 to determine whether the rotation path is deviated. The force plate 30 is connected to the analysis device 10 and placed under the athlete's 100 foot area to measure the pedaling force, so that the athlete 100 can measure the pedaling force with each step. The force plate 30 will generate a pressure signal and transmit it to the analysis device 10. The analysis device 10 compares the rotation path line M1 with the displacement path of the axial foot under continuous force, thereby determining whether the rotation path is deviated. By mutual correction of the two determination methods and improvement of limb movements, the best pedaling pace and the rotation path line M1 are further found, thereby conducting targeted training in the direction of rotation displacement to improve the competition results of the chain ball event.

On the other hand, the analysis device 10 provides an input of the throwing distance of the chain ball 101 each time the athlete 100 performs the chain ball project, and annotates the throwing distance of the chain ball 101 in the first curve chart S01. The chart data of each training is sorted and stored according to date and time, and can be transmitted to the cloud database 70 for storage and application. Not only can the training history of the athlete 100 be automatically established, but a large amount of data can also be collected for big data analysis, thereby analyzing whether the training method of the athlete 100 is suitable and understanding the degree of improvement. Furthermore, the corresponding time axes of the first curve graph S01 and the second curve graph S02 are embedded in the images captured by the drone 20 and the side camera 40. When the analysis device 10 clicks on the curve position on the first curve graph S01 or the second curve graph S02, the analysis device 10 starts playing the captured images with the corresponding time axis, and then cross-compares the real response levels of the first curve graph S01 and the second curve graph S02, so that the athlete 100 can more intuitively understand the limb movements that need to be improved, so as to effectively improve the training effect.

However, what is described above is only a preferred embodiment of the present invention and should not be used to limit the scope of implementation of the present invention; that is, all equivalent changes and modifications made within the scope of the patent application of the present invention should still fall within the scope of coverage of the patent of the present invention.

[The present invention] 10: Analysis device 20: Air-shot machine 30: Force plate 40: Side-shot machine 50: Grip gloves 60: Back-shot machine 70: Cloud database 100: Athlete 101: Chain ball 102: Limb joint 103: Head A1: Start time point A2, A3, A4: Force positioning point A5: End time point S01: First curve graph S02: Second curve graph M1: Rotation path line L: Height θ: Angle

FIG1 is a schematic diagram of the component blocks of the present invention. FIG2 is a schematic diagram of the aerial view of the present invention. FIG3 is a schematic diagram of the present invention displaying the rotation path with an aerial view. FIG4 is a schematic diagram of the present invention analyzing the lower limb body data with a side-shot image. FIG5 is a schematic diagram of the present invention analyzing the upper limb body data with a side-shot image. FIG6 is a curve diagram of the first curve chart of the present invention. FIG7 is a curve diagram of the angle between the arm and the vertical axis displayed in the second curve chart of the present invention. FIG8 is a curve diagram of the height of the shoulder and wrist displayed in the second curve chart of the present invention. FIG9 is a curve diagram (I) of the present invention combining the first curve chart and the second curve chart. Figure 10 is a curve diagram of the present invention that combines the first curve diagram and the second curve diagram (Part 2)

10:Analysis device

20: Drone

30: Force plate

40: Side camera

50: Grip gloves

60: Rear camera

70: Cloud database

Claims (10)

An integrated sensing and analysis system for the track and field chain ball event includes: an analysis device, which is used to integrate filming data and measurement data, and the analysis device establishes a common starting time point and an ending time point when starting measurement; an aerial camera, which is connected to the analysis device and shoots images above the athlete's head, and the analysis device uses an image recognition method to identify the relative position of the athlete's head and the chain ball, and the analysis device analyzes the movement trajectory and speed of the chain ball with the athlete's head as the center, and further establishes a first curve chart of the chain ball speed; A force plate, which is connected to the analysis device and placed under the athlete's feet to measure the pedaling force. The analysis device determines the axial foot that is continuously subjected to force and the force-applying foot that is instantly subjected to force, and further establishes the force-applying positioning point for the athlete to complete each rotation; A side camera, which is connected to the analysis device and set up on the athlete's side to take images. The analysis device uses the backbone identification method to identify the relative positions of the athlete's limb joints, and further establishes the force-applying positioning point for the athlete to complete each rotation. In one step, the analyzed shoulder height, arm elevation angle, hip joint height, knee angle, arm elevation angle and trunk angle are used to establish multiple second curve graphs; the analysis device inserts the starting time point, the ending time point and the force positioning point of each rotation into the first curve graph and the second curve graph, so that the problem points that need to be adjusted can be quickly found through the graph, thereby accurately adjusting the athlete's limb movements and training methods for the chain ball project. An integrated sensing and analysis system for track and field chain ball as described in claim 1, wherein the analysis device is a set of overlapping graphs of the first curve graph and at least one of the second curve graphs, so as to intuitively determine the corresponding position on the second curve graph where a significant change in chain ball speed occurs, thereby improving the athlete's limb movements in a targeted manner. An integrated sensing and analysis system for the athletics chain ball project as described in claim 2, wherein the corresponding time axes of the first curve graph and the second curve graph are embedded in the images captured by the drone and the side camera, so that when the analysis device clicks on a curve position on the first curve graph or the second curve graph, the analysis device starts playing the captured image with the corresponding time axis. An integrated sensing and analysis system for the track and field chain ball project as described in claim 1, wherein the analysis device records the head position of each athlete within the range of the start time point and the end time point, and draws a rotation path line. The rotation path line can be compared with the throwing direction of the chain ball to determine whether the rotation path is deviated. An integrated sensing and analysis system for track and field chain ball as described in claim 4, wherein the analysis device is connected to a grip glove, the grip glove is put on the athlete's hand and is used to withstand the centrifugal force of the chain ball, the analysis device analyzes the grip change to determine whether the athlete's hands are stiff and whether the arm is bent, and the start time and end time of the analysis device are the start time and end time of the grip glove being subjected to force. An integrated sensing and analysis system for track and field tackle as described in claim 1, wherein the analysis device is connected to a rear camera, which is mounted behind the athlete to capture images. The analysis device uses a backbone recognition method to identify the relative positions of the athlete's limb joints, and the images captured by the rear camera compensate for the lack of field of view of the side camera, thereby improving the accuracy of the second curve chart. An integrated sensing and analysis system for the track and field chain shot event as described in claim 1, wherein the analysis device provides input of the chain shot throwing distance each time the athlete performs the chain shot event, and the chain shot throwing distance is noted in the first curve chart, and the chart data of each training session is stored in order by date and time, thereby analyzing whether the athlete's training method is suitable and understanding the extent of improvement. An integrated sensing and analysis system for the athletics telescopic ball project as described in claim 7, wherein the analysis device is connected to a cloud database, and the analysis device uploads all of the first curve graph and the second curve graph to the cloud database, and the cloud database performs big data analysis, and uses the big data of the cloud database to determine the abnormal curve, and then a reminder is marked for the abnormal curve portion. The integrated sensing and analysis system for the track and field chain ball project as described in claim 1, wherein the step of obtaining the first curve chart by an image recognition method includes: a. recording the position of the chain ball in each frame for marking; b. calculating the change in the position of the chain ball in the previous and next frames to obtain the chain ball displacement; c. obtaining the chain ball speed by dividing the displacement of the speed formula by time; d. using the moving average method to eliminate the jagged edges of the original speed curve; e. recording the athlete information and speed data into the analysis device. The integrated sensing and analysis system for the track and field tache ball project as described in claim 1, wherein the step of obtaining the second curve chart by the skeleton recognition method includes: a. using a heat map to detect the body position of the athlete in the image; b. performing pixel analysis based on the body position to find all limb nodes; c. marking the positions of all limb nodes in each frame; d. calculating the height curve and angle curve using the corresponding limb nodes; e. using the moving average method to eliminate the jagged edges of the height curve or the angle curve; and f. recording the athlete information, height data, and angle data into the analysis device.

Images