A speed roller skating motion trajectory tracking and video acquisition method and device

Abstract

The present invention relates to the technical field of video tracking, and discloses a method and device for tracking and video capturing of a speed roller skating motion trajectory. The method and device comprise the following steps: dividing a sports field into sub-fields, planning a critical area in the sub-fields, determining whether to execute a tracking and video capturing switching mode, extracting comprehensive motion data from a sub-trajectory video, generating a remark sub-video, and sequentially numbering the remark sub-videos and splicing them head to tail to generate a speed roller skating motion trajectory video. Compared with the prior art, the present invention can continuously and accurately track and capture the motion trajectory, avoid interruptions and delays in tracking and video capturing of the motion trajectory, and can obtain a motion trajectory video that can completely represent and analyze the motion trajectory of a training personnel's speed roller skating by sequentially numbering the remark sub-videos and splicing them head to tail, and can ensure a seamless splicing effect between adjacent remark sub-videos.

Description

Technical Field

[0001] The present invention relates to the field of video tracking technology, and more specifically, to a speed roller skating motion trajectory tracking and video acquisition method and device. Background Art

[0002] Speed ​​skating is a competitive sport that requires extremely high skills, speed and stability of athletes. With the rapid development of computer vision and sensor technology, in speed skating training, it is necessary to use computer vision and sensor technology to track and capture the speed skating trajectory of trainees, so as to obtain real-time video of trainees' speed skating, and provide an accurate basis for subsequent speed skating analysis and targeted training.

[0003] The patent application with reference publication number CN115696024A discloses a method and device for tracking personnel targets in a smart park based on video surveillance. The monitoring video acquisition equipment set in the smart park is used to identify the personnel targets in the park and generate tracking trajectories. Finally, all the tracking trajectories are combined to form a complete tracking trajectory for storage, so that the trajectory tracking of strangers in the smart park can be realized, which reduces the potential safety hazards in the smart park to a certain extent.

[0004] Existing motion trajectory tracking and video acquisition methods use several independent cameras to continuously track and capture videos of the motion trajectory of a person, thereby achieving effective operation of segmenting and then integrating the motion trajectory of the person. For example, in the above-mentioned patent application, the tracking trajectory of the person is obtained by monitoring the video acquisition equipment, and all the tracking trajectories are combined to form a complete tracking trajectory. Although this method can achieve continuous tracking of the motion trajectory of the person, in the field training of speed skating, due to the fast movement speed of the trainees, the distance and duration between two adjacent motion trajectories that can be provided to the camera for tracking switching are short, which makes it easy for two adjacent cameras to have delays or interruptions at the connection of the motion trajectories when capturing the motion trajectory video, which leads to inaccurate tracking and video acquisition operations of the trainees' motion trajectories, and the accuracy of subsequent motion trajectory splicing cannot be ensured.

[0005] In view of this, the present invention proposes a speed skating motion trajectory tracking and video acquisition method and device to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a speed roller skating motion trajectory tracking and video acquisition method, applied to a video controller, comprising:

[0007] S1: Filtering the overhead image of the sports venue from the database, and dividing the sports venue into adjacent sub-venues based on the venue division criteria;

[0008] S2: Obtain the comprehensive site data of the sub-site, which includes the critical curvature, linear area ratio and boundary span distance, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude;

[0009] S3: obtaining the real-time position of the training personnel in the sub-venue, and determining whether to execute the tracking and video acquisition switching mode based on the area switching criteria;

[0010] S4: if the tracking and video acquisition switching mode is executed, the sub-track video of the training personnel is obtained, and the comprehensive motion data is extracted from the sub-track video, the comprehensive motion data including the center offset value, the average motion speed value and the center of gravity sway value, and the sub-track video is annotated to generate an annotated sub-video;

[0011] S5: The annotated sub-videos are numbered in sequence and spliced ​​head to tail to generate a speed roller skating motion trajectory video of the training personnel.

[0012] Furthermore, the site division criterion is: the dividing line between two adjacent sub-sites is located at the intersection of the extended lines of the coverage ranges of the shooting angles of two adjacent cameras;

[0013] The sub-site division method is as follows:

[0014] In the overhead image of the sports field, lines are drawn along the positions of the outer edge and the inner edge of the sports field to obtain an outer boundary line and an inner boundary line;

[0015] Draw extension lines from both sides of the lens of A cameras to intersect with the outer boundary line, and record the range between the two extension lines as the shooting angle coverage;

[0016] The intersection of the extended lines of the coverage ranges of two adjacent shooting angles is recorded as the extended intersection point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the extended intersection point to obtain A segmentation lines;

[0017] Using A segmentation lines as segmentation criteria, the overhead image of the sports field is divided into A adjacent sub-images, and the sports fields corresponding to the A sub-images are recorded as sub-fields.

[0018] Furthermore, the critical curvature is obtained as follows:

[0019] Scan the sub-images of A sub-fields to obtain A grayscale images, and count the number of pixel points located on the outer boundary line and the inner boundary line one by one to obtain A outer image values ​​and A inner image values;

[0020] Compare the A external image values ​​with the sum of the A external image values ​​and the A internal image values, respectively, to obtain A critical curvatures;

[0021] The expression of critical curvature is:

[0022] ;

[0023] In the formula, For the The critical curvature of each sub-site, =1,2...A, For the The external image value of each sub-site, For the The intrinsic image value of each sub-site.

[0024] Furthermore, the method for obtaining the linear region ratio is:

[0025] Draw lines along the outer edges of the grayscale images of the A sub-fields that do not belong to the outer boundary line and the inner boundary line to obtain two side lines;

[0026] Take one of the side lines as the standard line, draw two moving lines parallel to the standard line in the sub-field, and constantly adjust the distance between the two moving lines;

[0027] The number of pixel points on the outer boundary line and the inner boundary line between the two moving lines is counted respectively, and recorded as a first value and a second value. When the first value is equal to the second value, the two moving lines are stopped from being adjusted, and the outer boundary line, the inner boundary line and the area between the two moving lines are recorded as a linear area.

[0028] Count the number of all pixels in the grayscale image of A sub-sites and the number of pixels in the linear region respectively, and obtain A total pixel value and A total linear value;

[0029] After comparing A linear total values ​​with the corresponding A pixel total values, A linear area ratios are obtained;

[0030] The expression of the linear region ratio is:

[0031] ;

[0032] In the formula, For the The linear area ratio of the sub-sites, For the The linear total value of the sub-sites, For the The total pixel value of each sub-site;

[0033] The expression of site buffer coefficient is:

[0034] ;

[0035] In the formula, For the The site buffer coefficient of each sub-site, For the The boundary span distance of each sub-site, , , All of them are proportional coefficients greater than 0.

[0036] Furthermore, the critical area is divided as follows:

[0037] After multiplying the site buffer coefficients of A sub-sites by the standard area amplitude one by one, A area buffer values ​​are obtained;

[0038] The expression for the area buffer value is:

[0039] ;

[0040] In the formula, For the The area buffer value of each sub-site, is the standard area amplitude;

[0041] In the In the sub-fields, the distance The sub-site The position corresponding to the length of the buffer value of the region is taken as the critical point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the critical point to obtain A critical lines;

[0042] The first The critical line to sub-sites, and the area between the outer boundary line and the inner boundary line is recorded as the critical area, and A critical areas are obtained.

[0043] Furthermore, the region switching criterion is: when the switching point touches the trigger line, the switching operation is performed;

[0044] The method for determining whether to execute the tracking and video acquisition switching mode is as follows:

[0045] The motion image of the trainee in the current sub-field is captured by a camera, a closed line is drawn along the position of the outer edge of the trainee in the motion image, and the area within the closed line is recorded as the posture area;

[0046] Mark the four corner points of the critical area of ​​the current sub-site, draw two diagonal lines of the critical area of ​​the current sub-site with the four corner points as base points, and record the intersection of the two diagonal lines as the trigger point;

[0047] An auxiliary line perpendicular to the outer boundary line and the inner boundary line is drawn through the trigger point to obtain the trigger line, and B switching points are equidistantly marked on the outer boundary of the body area based on a preset marking length, and the relative positions of the B switching points and the trigger line are detected;

[0048] When there is no switching point touching the auxiliary line, it is determined that the tracking and video acquisition switching modes are not executed;

[0049] When a switching point touches the auxiliary line, it is determined to execute the tracking and video acquisition switching mode.

[0050] Furthermore, the method for extracting the center offset value is:

[0051] Count the motion durations of the A sub-trajectory videos one by one to obtain A motion durations, and record the sub-trajectory video corresponding to the minimum motion duration as the target video;

[0052] Mark out at preset time intervals At a certain time point, the target video is captured. The image at each time point is obtained A moment image;

[0053] After dividing the remaining A-1 exercise durations by the minimum value of the exercise duration, A-1 interval multiples are obtained, and the A-1 interval multiples are multiplied by the preset duration interval to obtain A-1 revised duration intervals;

[0054] Take the A-1 revised time intervals as the standard and mark the remaining A-1 sub-track videos. At a certain time point, A-1 sub-trajectory videos are captured. An image at a certain point in time;

[0055] exist The positions of the two feet of the trainee are respectively identified in the image at each moment, two foot regions are obtained, and all the pixels in the two foot regions are marked;

[0056] The distance values ​​from the pixel points in one of the foot regions to the pixel points in the other foot region are measured one by one by using a scale, and the two pixel points corresponding to the minimum value of the distance value are recorded as the first pixel point and the second pixel point;

[0057] The first pixel point and the second pixel point are connected to form a pixel line, and the minimum distance between the midpoint of the pixel line and the preset center line is measured to obtain Individual deviation value;

[0058] Will The sub-deviation values ​​are accumulated and averaged to obtain A center offset values;

[0059] The expression for the center offset value is:

[0060] ;

[0061] In the formula, For the The center offset value of the sub-track video, For the The sub-track video The sub-deviation value.

[0062] Furthermore, the method for extracting the average speed value of motion is:

[0063] In the A sub-track videos, the speed sensor detects the trainee’s The instantaneous speed at a certain time point is obtained Speed ​​of individual;

[0064] After removing the maximum and minimum sub-speeds, the remaining The speeds of each individual are accumulated and averaged to obtain A average speed values ​​of the movements;

[0065] The expression of the average speed of motion is:

[0066] ;

[0067] In the formula, For the The average speed of the sub-trajectory video, For the The sub-track video Speed ​​of the child.

[0068] Furthermore, the method for obtaining the center of gravity sway value is:

[0069] exist In each moment image, mark the position of the trainee's head one by one, obtain the head area, and mark all the pixels in the head area;

[0070] The distance values ​​from all the pixels in the head area to the ground of the sub-site are measured one by one using a scale to obtain The center of gravity distance values ​​are obtained by subtracting the maximum value of the center of gravity distance value from the minimum value of the center of gravity distance value to obtain A center of gravity sway values;

[0071] The expression of the center of gravity sway value is:

[0072] ;

[0073] In the formula, For the The center of gravity shake value of each sub-track video, For the The maximum value of the center of gravity distance of the sub-trajectory videos, For the The minimum value of the center-of-gravity distance of the sub-trajectory videos;

[0074] The method for generating the annotated sub-video is:

[0075] The center offset value, average speed value and center of gravity shaking value of A sub-track videos are respectively compared with the offset safety value , mean safety value and sway safety value Compare;

[0076] when Greater than When The center offset value of each sub-track video is recorded as the annotation data;

[0077] when Less than When The average speed of each sub-trajectory video is recorded as the remark data;

[0078] when Greater than When The center of gravity shaking value of each sub-trajectory video is recorded as the remark data;

[0079] Remark boxes are constructed on the A sub-track videos respectively, and the remark data are imported into the remark boxes of the corresponding sub-track videos one by one to generate A remark sub-videos.

[0080] A motion trajectory tracking and video acquisition device, the device includes a video controller, the video controller integrates a site division module, an area planning module, a mode determination module, a video remark module and a video splicing module, the site division module, the area planning module, the mode determination module, the video remark module and the video splicing module are used together to realize a speed roller skating motion trajectory tracking and video acquisition method, wherein:

[0081] A venue division module is used to filter out the overhead image of the sports venue from the database, and divide the sports venue into adjacent sub-venues based on the venue division criteria;

[0082] The regional planning module is used to obtain the comprehensive site data of the sub-site, including the critical curvature, linear area ratio and boundary span distance, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude;

[0083] A mode determination module is used to obtain the real-time position of the training personnel in the sub-venue and determine whether to execute the tracking and video acquisition switching mode based on the area switching criteria;

[0084] The video annotation module is used to obtain the sub-track video of the trainee, extract the comprehensive motion data from the sub-track video, the comprehensive motion data includes the center offset value, the average speed value and the center of gravity sway value, and annotate the sub-track video to generate an annotated sub-video;

[0085] The video splicing module is used to sequentially number the annotated sub-videos and splice them head to tail to generate a speed roller skating motion trajectory video of the training personnel.

[0086] The technical effects and advantages of the speed skating motion trajectory tracking and video acquisition method and device of the present invention are as follows:

[0087] The present invention divides a sports field into adjacent sub-fields, and can divide a sports field with a larger overall area and span into smaller sub-fields as needed, thereby reducing the workload of the camera in motion trajectory tracking and video acquisition. By acquiring comprehensive field data of the sub-fields, the field buffer coefficient is calculated, and the critical area is planned in the sub-field in combination with the field buffer coefficient and the standard area amplitude, and the working area to be turned on of the camera can be marked between two adjacent sub-fields, so that when the training personnel cross between two adjacent sub-fields, the motion trajectory can be tracked and the video can be captured continuously and accurately, avoiding the interruption and delay of motion trajectory tracking and video acquisition. The delay phenomenon can be eliminated, and the comprehensive motion data of the sub-track video can be used to generate an annotation sub-video from the sub-track video, so that all-round data collection can be carried out on the training personnel's motion performance in the sub-venues, and the specific type of motion trajectory of each sub-venue can be accurately analyzed. Finally, by numbering the annotation sub-videos in sequence and splicing them head to tail, a motion trajectory video that can completely represent and analyze the motion trajectory of the training personnel's speed skating can be obtained, which can not only ensure the seamless splicing effect between adjacent annotation sub-videos, but also provide accurate video data for the subsequent targeted training of speed skating of the training personnel, thereby helping the training personnel to better conduct speed skating training. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 A schematic diagram of a flow chart of a speed skating motion trajectory tracking and video acquisition method provided in the first embodiment of the present invention;

[0089] Figure 2 A schematic diagram of the sub-sites divided according to the first embodiment of the present invention;

[0090] Figure 3 A schematic diagram of a device provided in Embodiment 2 of the present invention;

[0091] Figure 4 A schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention;

[0092] Figure 5 A schematic diagram of the structure of a computer-readable storage medium provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0093] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0094] Example 1: Please refer to Figure 1 and Figure 2 As shown, the speed skating motion trajectory tracking and video acquisition method described in this embodiment is applied to a video controller, including:

[0095] S1: Filtering the overhead image of the sports venue from the database, and dividing the sports venue into adjacent sub-venues based on the venue division criteria;

[0096] Bird's-eye view images refer to images taken from a high altitude and containing the panoramic view of the sports venue and the distribution positions of cameras. They serve as the image basis for the subsequent division of the sports venue. Therefore, it is necessary to filter out bird's-eye view images that can contain the panoramic view of the sports venue and the distribution positions of cameras from the database, so that the sports venue and cameras can be clearly observed from a bird's-eye view.

[0097] After the overhead image is acquired, the shape of the sports field in the overhead image is usually an ellipse. At this time, the overall span and area of ​​the sports field are large, so that the movement trajectory span generated by the trainer during speed skating training on the sports field is also long. In order to achieve more refined tracking and video acquisition effects of the movement trajectory, the sports field needs to be divided into multiple adjacent sub-fields with smaller movement trajectory spans;

[0098] When dividing the sports field into sub-fields, it is necessary to ensure that each sub-field can maintain accurate contact with the corresponding camera within the constraints of the field division criteria;

[0099] The site division criterion is: the dividing line between two adjacent sub-sites is located at the intersection of the extended lines of the coverage range of the shooting angles of two adjacent cameras; this ensures the rationality and consistency of the division positions between two adjacent sub-sites, and ensures that the basis for the division of each sub-site is consistent;

[0100] The sub-site division method is as follows:

[0101] In the overhead image of the sports field, lines are drawn along the positions of the outer edge and the inner edge of the sports field to obtain an outer boundary line and an inner boundary line;

[0102] Draw extension lines from both sides of the lens of A cameras and intersect with the outer boundary line, and record the range between the two extension lines as the shooting angle coverage range; the shooting angle coverage range of each camera is a fan-shaped structure, and the entire sports venue can be fully covered by overlapping and splicing the shooting angle coverage ranges of multiple cameras;

[0103] The intersection of the extended lines of the coverage ranges of two adjacent shooting angles is recorded as the extended intersection point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the extended intersection point to obtain A segmentation lines;

[0104] Using A segmentation lines as segmentation criteria, the overhead image of the sports field is divided into A adjacent sub-images, and the sports fields corresponding to the A sub-images are recorded as sub-fields.

[0105] Based on the above sub-venue division method, the sports field is divided into sub-venues for illustration. For details, please refer to Figure 2 As shown in the figure, YCD is the sports field, ZCD is the sub-field, XS is the camera, YC is the extension line of the shooting angle coverage range, FG is the dividing line, WB is the outer boundary line, and NB is the inner boundary line.

[0106] It should be noted that, due to the differences in camera performance and the actual shapes of sports venues, the overall shape structure of each sub-venue may be different, so the area and orientation of the sub-venue may also be different.

[0107] S2: Obtain the comprehensive site data of the sub-site, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude; the comprehensive site data includes critical curvature, linear area ratio and boundary span distance;

[0108] Comprehensive venue data refers to multi-dimensional data that can represent inherent information such as the shape and size of the sub-venue, so as to be able to represent the performance of each sub-venue in speed skating in many aspects and provide a data basis for the calculation of the subsequent venue buffer coefficient;

[0109] Comprehensive site data include critical curvature, linear area ratio, and boundary span distance;

[0110] The critical curvature refers to the curvature of the boundary contour in the sub-image of the sub-site, which can represent the curvature performance of the sub-site. The larger the critical curvature, the greater the curvature of the boundary contour of the sub-site, and the greater the site buffer coefficient.

[0111] The method for obtaining the critical curvature is:

[0112] Scan the sub-images of A sub-fields to obtain A grayscale images, and count the number of pixels located on the outer boundary line and the inner boundary line one by one to obtain A outer image values ​​and A inner image values; when the number of pixels on a certain boundary line is more, it means that the length of the boundary line is longer. Since the area size of the location of each pixel point is the same, the longer the length of the boundary line, the greater the curvature of the boundary line;

[0113] Compare the A external image values ​​with the sum of the A external image values ​​and the A internal image values, respectively, to obtain A critical curvatures;

[0114] The expression of critical curvature is:

[0115] ;

[0116] In the formula, For the The critical curvature of each sub-site, =1,2...A, For the The external image value of each sub-site, For the The intrinsic image value of each sub-site.

[0117] The linear area ratio refers to the ratio of the area of ​​the linear area in the sub-image of the sub-site to the total area, which can be used to represent the area of ​​the linear area of ​​the sub-site. The larger the linear area ratio, the larger the area of ​​the linear area of ​​the sub-site, and the smaller the site buffer coefficient.

[0118] The method for obtaining the linear region ratio is:

[0119] Draw lines along the outer edges of the grayscale images of the A sub-fields that do not belong to the outer boundary line and the inner boundary line to obtain two side lines;

[0120] Take one of the side lines as the standard line, draw two moving lines parallel to the standard line in the sub-field, and constantly adjust the distance between the two moving lines;

[0121] The number of pixel points on the outer boundary line and the inner boundary line between the two moving lines is counted respectively, and recorded as the first value and the second value. When the first value is equal to the second value, the two moving lines are stopped from being adjusted, and the outer boundary line, the inner boundary line and the area between the two moving lines are recorded as the linear area. The number of linear areas in the grayscale image of each sub-field may be 0, may be only 1, or may be more than one, which is determined by the specific shape of the sub-field.

[0122] Count the number of all pixels in the grayscale image of A sub-sites and the number of pixels in the linear region respectively, and obtain A total pixel value and A total linear value;

[0123] After comparing A linear total values ​​with the corresponding A pixel total values, A linear area ratios are obtained;

[0124] The expression of the linear region ratio is:

[0125] ;

[0126] In the formula, For the The linear area ratio of the sub-sites, For the The linear total value of the sub-sites, For the The total pixel value of each sub-field.

[0127] The boundary span distance refers to the distance between the points at the two diagonal positions in the sub-image of the sub-site, which can be used to represent the site span size of the sub-site. When the boundary span distance is larger, it means that the site span of the sub-site is larger, and the site buffer coefficient is smaller. When obtaining the boundary span distance, it is necessary to measure the diagonal lengths of the two diagonals in the grayscale images of A sub-sites one by one through a scale, and then take the maximum value of the diagonal lengths as the boundary span distance.

[0128] When the comprehensive site data is obtained, the site buffer coefficient can be calculated based on the comprehensive site data, so that the site buffer coefficient can be used as the basis for subsequent combination with the standard area amplitude, thereby obtaining the data basis for planning the critical area;

[0129] The expression of site buffer coefficient is:

[0130] ;

[0131] In the formula, For the The site buffer coefficient of each sub-site, For the The boundary span distance of each sub-site, , , All of them are proportional coefficients greater than 0;

[0132] in, , , , The setting is to balance the proportion of critical curvature, linear area ratio and boundary span distance in the site buffer coefficient, so as to ensure that the changes in the critical curvature, linear area ratio and boundary span distance can cause corresponding changes in the site buffer coefficient.

[0133] The critical area is the transitional adjacent area between each sub-site and the next sub-site. It can also provide a basis for judging and analyzing whether the camera corresponding to the next sub-site is activated or not, thereby ensuring the accurate tracking of the motion trajectory and video acquisition in each sub-site.

[0134] The critical area is divided as follows:

[0135] After multiplying the site buffer coefficients of A sub-sites by the standard area amplitude one by one, A area buffer values ​​are obtained; the standard area amplitude is used to standardize the length of the motion direction corresponding to the critical area, which can provide a numerical basis for the length of the motion direction corresponding to the critical area, and by increasing or decreasing the standard area amplitude, the adaptive change of the size of the motion direction of the critical area of ​​each sub-site is achieved;

[0136] The expression for the area buffer value is:

[0137] ;

[0138] In the formula, For the The area buffer value of each sub-site, is the standard area amplitude;

[0139] In the In the sub-fields, the distance The sub-site The position corresponding to the length of the buffer value of the region is taken as the critical point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the critical point to obtain A critical lines;

[0140] The first The critical line to sub-sites, and the area between the outer boundary line and the inner boundary line is recorded as the critical area, and A critical areas are obtained.

[0141] It should be noted that the shape of the critical area is not fixed and unique. It may be a rectangle, an arc, or a mixture of rectangle and arc. The actual shape of the critical area is related to the actual shape of the sub-site where the critical area is located and the next sub-site adjacent to it.

[0142] S3: obtaining the real-time position of the training personnel in the sub-venue, and determining whether to execute the tracking and video acquisition switching mode based on the area switching criteria;

[0143] The real-time position refers to the real-time position of the trainer when he / she is speed skating in the sub-venue. By collecting and analyzing the real-time position, the real-time movement trajectory of the trainer in each sub-venue can be determined, thereby providing support for the subsequent switching of tracking and video acquisition;

[0144] The tracking and video acquisition switching mode is used to switch the corresponding cameras for tracking and video acquisition of the trainees' motion trajectory in each sub-venue, so as to ensure that the cameras in each sub-venue can accurately track and capture the motion trajectory of the trainees in each sub-venue. In determining whether to execute the tracking and video acquisition switching mode, it is necessary to provide restriction conditions for the mode switching under the restriction of the regional switching criteria;

[0145] The area switching rule is: when the switching point touches the trigger line, the switching operation is executed; this ensures that the camera of the next sub-site can track and capture the movement trajectory of the trainee in the first place, avoiding the phenomenon of tracking and video capture delay or loss;

[0146] The method for determining whether to execute the tracking and video acquisition switching mode is as follows:

[0147] The motion image of the trainee in the current sub-field is captured by a camera, a closed line is drawn along the position of the outer edge of the trainee in the motion image, and the area within the closed line is recorded as the body area; the body area is used as a representation of the body coverage area of ​​the trainee in the motion image, and can provide area restrictions for the marking of subsequent switching points;

[0148] Mark the four corner points of the critical area of ​​the current sub-site, draw two diagonal lines of the critical area of ​​the current sub-site with the four corner points as base points, and record the intersection of the two diagonal lines as the trigger point;

[0149] An auxiliary line perpendicular to the outer boundary line and the inner boundary line is drawn through the trigger point to obtain the trigger line. B switching points are equidistantly marked on the outer boundary of the body area with a preset mark length as a standard, and the relative positions of the B switching points and the trigger line are detected; the preset mark length is used to represent the distance between two adjacent switching points, which can ensure the independence of each switching point, thereby avoiding the phenomenon that two adjacent switching points are too close or too far away from each other;

[0150] When there is no switching point touching the auxiliary line, the movement trajectory of the trainee has not yet reached the range where the camera of the next sub-field starts tracking and video acquisition, and it is determined that the switching mode of tracking and video acquisition is not executed;

[0151] When there is a switching point that touches the auxiliary line, the movement trajectory of the trainee reaches the range where the camera of the next sub-field starts tracking and video acquisition, and it is determined to execute the tracking and video acquisition switching mode.

[0152] It should be noted that when the tracking and video acquisition switching mode is executed, the camera located in the next sub-field of the current sub-field will be immediately started and be on standby at all times. When the trainee enters the sub-field corresponding to the camera of the next sub-field, the video image within the coverage range of the corresponding camera angle of the camera will be captured. In this way, when the trainee's movement trajectory reaches the coverage range of the camera angle, the trainee's movement trajectory video can be seamlessly captured.

[0153] S4: Obtain the sub-trajectory video of the training personnel, extract comprehensive motion data from the sub-trajectory video, the comprehensive motion data including the center offset value, the average motion speed value and the center of gravity sway value, and annotate the sub-trajectory video to generate an annotated sub-video;

[0154] The sub-track video refers to the motion track video of the trainees performing speed skating in the sub-venue captured by the camera, which can serve as the basis for subsequent overall video acquisition and analysis of the trainees' speed skating. The method for acquiring the sub-track video only needs to be realized through the camera video acquisition corresponding to each sub-venue, thereby obtaining A sub-track videos.

[0155] After obtaining the sub-track video, it is necessary to extract comprehensive motion data from the sub-track video so that the comprehensive motion data can provide a multi-dimensional numerical representation of the speed skating operation of the trainee in the sub-track video and provide a basis for the subsequent judgment and analysis of the sub-track video;

[0156] Comprehensive motion data includes center offset value, average motion speed value and center of gravity sway value;

[0157] The center offset value refers to the offset between the center point of the trainer's motion trajectory in the sub-trajectory video and the center line of the sub-field, which can represent the change of the center point of the trainer's motion trajectory. The larger the center offset value, the larger the offset between the center point of the trainer's motion trajectory in the sub-trajectory video and the center area of ​​the sub-field.

[0158] The method for extracting the center offset value is:

[0159] Count the motion durations of the A sub-trajectory videos one by one to obtain A motion durations, and record the sub-trajectory video corresponding to the minimum motion duration as the target video;

[0160] Mark out at preset time intervals At a certain time point, the target video is captured. The image at each time point is obtained The preset time interval is used to represent the time span between two adjacent time points, so as to ensure the independence of each time point and ensure that the body posture of the trainee in the time image corresponding to the time point will change, and the change of the body posture can be effectively collected and identified;

[0161] After dividing the remaining A-1 exercise durations by the minimum value of the exercise duration, A-1 interval multiples are obtained, and the A-1 interval multiples are multiplied by the preset duration interval to obtain A-1 revised duration intervals;

[0162] Take the A-1 revised time intervals as the standard and mark the remaining A-1 sub-track videos. At a certain time point, A-1 sub-trajectory videos are captured. An image at a certain point in time;

[0163] exist The positions of the two feet of the trainee are respectively identified in the image at each moment, two foot regions are obtained, and all the pixels in the two foot regions are marked;

[0164] The distance values ​​from the pixel points in one of the foot regions to the pixel points in the other foot region are measured one by one by using a scale, and the two pixel points corresponding to the minimum value of the distance value are recorded as the first pixel point and the second pixel point;

[0165] The first pixel point and the second pixel point are connected to form a pixel line, and the minimum distance between the midpoint of the pixel line and the preset center line is measured to obtain The preset center line refers to the route corresponding to the optimal motion trajectory of the trainee when performing speed skating in each sub-field, which provides a basis for the calculation of the subsequent center offset value and can determine the difference between the center of the trainee and the preset center line;

[0166] Will The sub-deviation values ​​are accumulated and averaged to obtain A center offset values;

[0167] The expression for the center offset value is:

[0168] ;

[0169] In the formula, For the The center offset value of the sub-track video, For the The sub-track video The sub-deviation value.

[0170] The average speed value refers to the average speed of the trainee's speed skating in the sub-venue in the sub-track video, which can be used to represent the average speed of the trainee's skating in the sub-venue. The larger the average speed value, the larger the average speed of the trainee's speed skating in the sub-venue in the sub-track video.

[0171] The method for extracting the average speed of movement is:

[0172] In the A sub-track videos, the speed sensor detects the trainee’s The instantaneous speed at a certain time point is obtained Speed ​​of individual;

[0173] After removing the maximum and minimum sub-speeds, the remaining The speeds of each individual are accumulated and averaged to obtain A average speed values ​​of the movements;

[0174] The expression of the average speed of motion is:

[0175] ;

[0176] In the formula, For the The average speed of the sub-trajectory video, For the The sub-track video Speed ​​of the child.

[0177] The center of gravity sway value refers to the maximum value of the up and down change range of the trainee's body corresponding to the height in the sub-track video, which can represent the change of the center of gravity of the trainee in the sub-track video. When the center of gravity sway value is larger, it means that the maximum value of the up and down change range of the trainee's body corresponding to the height in the sub-track video is larger;

[0178] The method for obtaining the center of gravity sway value is:

[0179] exist In each moment image, mark the position of the trainee's head one by one, obtain the head area, and mark all the pixels in the head area;

[0180] The distance values ​​from all the pixels in the head area to the ground of the sub-site are measured one by one using a scale to obtain The center of gravity distance values ​​are obtained by subtracting the maximum value of the center of gravity distance value from the minimum value of the center of gravity distance value to obtain A center of gravity sway values;

[0181] The expression of the center of gravity sway value is:

[0182] ;

[0183] In the formula, For the The center of gravity shake value of each sub-track video, For the The maximum value of the center of gravity distance of the sub-trajectory videos, For the The minimum value of the centroid distance of the sub-track videos.

[0184] After obtaining the comprehensive motion data, it is necessary to make relevant video annotations for each sub-track video, so as to explain and annotate the prominent features and real content of each sub-track video, so as to facilitate the subsequent rapid and accurate understanding and judgment of each sub-track video, and finally record the sub-track video with video annotations as the annotated sub-video;

[0185] The method for generating the annotated sub-video is:

[0186] The center offset value, average speed value and center of gravity shaking value of A sub-track videos are respectively compared with the offset safety value , mean safety value and sway safety value Compare;

[0187] when Greater than , explain The center offset value of the sub-track video is greater than the offset safety value. If the center deviation of the sub-track video exceeds the safe range, the The center offset value of each sub-track video is recorded as the remark data; the offset safety value refers to the minimum value of the center offset value when the sub-track video is identified as the remark data, and is used as the basis for judging whether the center offset value is the remark data;

[0188] when Less than , explain The average speed of the sub-trajectory videos is less than the average safety value. If the average speed of the sub-track video does not reach the safe range, the The average speed value of each sub-track video is recorded as the remark data; the average safety value refers to the maximum value of the average speed value of the sub-track video when it is identified as the remark data, which is used as the basis for judging whether the average speed value of the motion is the remark data;

[0189] when Greater than , explain The center of gravity shaking value of the sub-track video is greater than the shaking safety value. If the shaking amplitude of the center of the sub-track video exceeds the safe range, the The center of gravity shaking value of each sub-track video is recorded as the remark data; the shaking safety value refers to the minimum value of the center of gravity shaking value when the sub-track video is identified as the remark data, which is used as the basis for judging whether the center of gravity shaking value is the remark data;

[0190] Construct a comment box on each of the A sub-track videos, and import the comment data into the comment box of the corresponding sub-track video one by one to generate A annotated sub-videos. The comment box is an information box used to provide comment data import for the sub-track video, so as to ensure that each annotated sub-video can be annotated with the corresponding comment data.

[0191] It should be noted that the number of annotation data in each annotated sub-video is between 1 and 3, and the number of annotation data in each annotated sub-video may vary, thereby reflecting the uniqueness of each annotated sub-video.

[0192] S5: numbering the annotated sub-videos in sequence and splicing them head to tail to generate a speed skating motion trajectory video of the training personnel;

[0193] After the annotated sub-videos are obtained, the A annotated sub-videos can only provide a video representation of the individual motion trajectory of the speed skating of the trainee in each sub-venue, and cannot provide an overall representation of the entire motion trajectory of the trainee in the sports venue. Therefore, the A annotated sub-videos need to be spliced ​​in order to obtain a speed skating motion trajectory video that represents the entire motion trajectory of the trainee in the sports venue as a whole, and enable the speed skating motion trajectory video to serve as a basis for subsequent analysis, judgment and guidance of the speed skating of the trainee;

[0194] In order to obtain the speed skating motion trajectory video, it is necessary to sequentially number A remark sub-videos. When performing the sequential numbering, the first sub-field that generates the sub-trajectory video is taken as the first number, and the sub-trajectory videos of each sub-field are sequentially numbered in ascending order according to the order in which the trainees speed skate through the sub-fields, and the numbers of the sub-trajectory videos are correspondingly transferred to the remark sub-videos, so that the sequential numbering of the remark sub-videos can be achieved. For example, when the first sub-field that generates the sub-trajectory video is 1, and the number of sub-fields is 12, the sub-trajectory videos corresponding to the sub-fields are numbered as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 according to the order in which the trainees speed skate through the sub-fields, and at the same time, the remark sub-videos corresponding to the sub-trajectory videos are also numbered as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12;

[0195] After the annotation sub-videos are numbered in sequence, all the annotation sub-videos are spliced ​​head to tail in the order of numbering from small to large, so that the end time of the previous annotation sub-video can be in a continuous state with the start time of the next sub-video on the timeline, thereby ensuring seamless splicing between two adjacent annotation sub-videos, and achieving an accurate and complete splicing effect of the speed roller skating motion trajectory of the trainee, and finally obtaining the required speed roller skating motion trajectory video;

[0196] In order to ensure that the end time of the previous annotation sub-video is continuous with the start time of the next sub-video on the timeline, the time when the trainee passes through each dividing line is used as the critical time of the start time and the end time.

[0197] In this embodiment, by dividing the sports field into adjacent sub-fields, the sports field with a larger overall area and span can be divided into smaller sub-fields as needed, thereby reducing the workload of the camera in motion trajectory tracking and video acquisition. By obtaining the comprehensive field data of the sub-fields, the field buffer coefficient is calculated, and the critical area is planned in the sub-field in combination with the field buffer coefficient and the standard area amplitude. The working area to be turned on by the camera can be marked between two adjacent sub-fields, so that when the training personnel cross between two adjacent sub-fields, the motion trajectory can be tracked and the video can be captured continuously and accurately, avoiding the interruption of motion trajectory tracking and video capture. and delay phenomenon, and through the comprehensive motion data of the sub-track video, the sub-track video can be generated into an annotation sub-video, so that all-round data collection can be carried out on the training personnel's sports performance in the sub-venue, and the specific type of motion trajectory of each sub-venue can be accurately analyzed. Finally, by numbering the annotation sub-videos in sequence and splicing them head to tail, a motion trajectory video that can completely represent and analyze the motion trajectory of the training personnel's speed skating can be obtained, which can not only ensure the seamless splicing effect between adjacent annotation sub-videos, but also provide accurate video data for the subsequent targeted training of speed skating of the training personnel, thereby helping the training personnel to better conduct speed skating training.

[0198] Example 2: Please refer to Figure 3 As shown, the part not described in detail in this embodiment is described in the first embodiment, and a motion trajectory tracking and video acquisition device is provided, which includes a video controller, and the video controller integrates a site division module, an area planning module, a mode determination module, a video remark module and a video splicing module. The site division module, the area planning module, the mode determination module, the video remark module and the video splicing module are used together to implement a speed skating motion trajectory tracking and video acquisition method, wherein:

[0199] A venue division module is used to filter out the overhead image of the sports venue from the database, and divide the sports venue into adjacent sub-venues based on the venue division criteria;

[0200] The regional planning module is used to obtain the comprehensive site data of the sub-site, including the critical curvature, linear area ratio and boundary span distance, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude;

[0201] A mode determination module is used to obtain the real-time position of the training personnel in the sub-venue and determine whether to execute the tracking and video acquisition switching mode based on the area switching criteria;

[0202] The video annotation module is used to obtain the sub-track video of the trainee, extract the comprehensive motion data from the sub-track video, the comprehensive motion data includes the center offset value, the average speed value and the center of gravity sway value, and annotate the sub-track video to generate an annotated sub-video;

[0203] The video splicing module is used to sequentially number the annotated sub-videos and splice them head to tail to generate a speed roller skating motion trajectory video of the training personnel.

[0204] Example 3: Please refer to Figure 4 As shown, this embodiment discloses an electronic device, including a processor and a memory;

[0205] Wherein, the memory stores a computer program that can be called by the processor;

[0206] The processor executes the speed roller skating motion trajectory tracking and video acquisition method by calling the computer program stored in the memory.

[0207] Since the electronic device introduced in this embodiment is an electronic device used to implement a speed roller skating motion trajectory tracking and video acquisition method in the first embodiment of the present application, based on the speed roller skating motion trajectory tracking and video acquisition method introduced in the embodiment of the present application, the technical personnel of the field can understand the specific implementation of the electronic device of the present embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is not described in detail here. As long as the technical personnel of the field implement the electronic device used in the speed roller skating motion trajectory tracking and video acquisition method in the embodiment of the present application, it belongs to the scope of protection of the present application.

[0208] Example 4: Please refer to Figure 5 As shown, this embodiment discloses a computer-readable storage medium on which a rewritable computer program is stored;

[0209] When the computer program is run, the speed roller skating motion trajectory tracking and video acquisition method is implemented.

[0210] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A speed skating motion trajectory tracking and video acquisition method, applied to a video controller, characterized in that: include: S1: Filtering the overhead image of the sports venue from the database, and dividing the sports venue into adjacent sub-venues based on the venue division criteria; S2: Obtain the comprehensive site data of the sub-site, which includes the critical curvature, linear area ratio and boundary span distance, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude; The method for obtaining the critical curvature is: Scan the sub-images of A sub-fields to obtain A grayscale images, and count the number of pixel points located on the outer boundary line and the inner boundary line one by one to obtain A outer image values ​​and A inner image values; Compare the A external image values ​​with the sum of the A external image values ​​and the A internal image values, respectively, to obtain A critical curvatures; The expression of critical curvature is: ； In the formula, For the The critical curvature of each sub-site, =1,2...A, For the The external image value of each sub-site, For the The intrinsic image value of each sub-site; The method for obtaining the linear region ratio is: Draw lines along the outer edges of the grayscale images of the A sub-fields that do not belong to the outer boundary line and the inner boundary line to obtain two side lines; Take one of the side lines as the standard line, draw two moving lines parallel to the standard line in the sub-field, and constantly adjust the distance between the two moving lines; The number of pixel points on the outer boundary line and the inner boundary line between the two moving lines is counted respectively, and recorded as a first value and a second value. When the first value is equal to the second value, the two moving lines are stopped from being adjusted, and the outer boundary line, the inner boundary line and the area between the two moving lines are recorded as a linear area. Count the number of all pixels in the grayscale image of A sub-sites and the number of pixels in the linear region respectively, and obtain A total pixel value and A total linear value; After comparing A linear total values ​​with the corresponding A pixel total values, A linear area ratios are obtained; The expression of the linear region ratio is: ； In the formula, For the The linear area ratio of the sub-sites, For the The linear total value of the sub-sites, For the The total pixel value of each sub-site; The expression of site buffer coefficient is: ； In the formula, For the The site buffer coefficient of each sub-site, For the The boundary span distance of each sub-site, , , All of them are proportional coefficients greater than 0; The critical area is divided as follows: After multiplying the site buffer coefficients of A sub-sites by the standard area amplitude one by one, A area buffer values ​​are obtained; The expression for the area buffer value is: ； In the formula, For the The area buffer value of each sub-site, is the standard area amplitude; In the In the sub-fields, the distance The sub-site The position corresponding to the length of the buffer value of the region is taken as the critical point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the critical point to obtain A critical lines; The first The critical line to sub-sites, and the area between the outer boundary line and the inner boundary line is recorded as the critical area, and A critical areas are obtained; S3: obtaining the real-time position of the training personnel in the sub-venue, and determining whether to execute the tracking and video acquisition switching mode based on the area switching criteria; The region switching criterion is: when the switching point touches the trigger line, the switching operation is performed; The method for determining whether to execute the tracking and video acquisition switching mode is as follows: The motion image of the trainee in the current sub-field is captured by a camera, a closed line is drawn along the position of the outer edge of the trainee in the motion image, and the area within the closed line is recorded as the posture area; Mark the four corner points of the critical area of ​​the current sub-site, draw two diagonal lines of the critical area of ​​the current sub-site with the four corner points as base points, and record the intersection of the two diagonal lines as the trigger point; An auxiliary line perpendicular to the outer boundary line and the inner boundary line is drawn through the trigger point to obtain the trigger line, and B switching points are equidistantly marked on the outer boundary of the body area based on a preset marking length, and the relative positions of the B switching points and the trigger line are detected; When there is no switching point touching the auxiliary line, it is determined that the tracking and video acquisition switching modes are not executed; When a switching point touches the auxiliary line, it is determined that the tracking and video acquisition switching modes are executed; S4: if the tracking and video acquisition switching mode is executed, the sub-track video of the training personnel is obtained, and the comprehensive motion data is extracted from the sub-track video, the comprehensive motion data including the center offset value, the average motion speed value and the center of gravity sway value, and the sub-track video is annotated to generate an annotated sub-video; S5: The annotated sub-videos are numbered in sequence and spliced ​​head to tail to generate a speed skating motion trajectory video of the training personnel.

2. A speed skating motion trajectory tracking and video acquisition method according to claim 1, characterized in that: The site division criterion is: the dividing line between two adjacent sub-sites is located at the intersection of the extended lines of the coverage ranges of the shooting angles of two adjacent cameras; The sub-site division method is as follows: In the overhead image of the sports field, lines are drawn along the positions of the outer edge and the inner edge of the sports field to obtain an outer boundary line and an inner boundary line; Draw extension lines from both sides of the lens of A cameras to intersect with the outer boundary line, and record the range between the two extension lines as the shooting angle coverage; The intersection of the extended lines of the coverage ranges of two adjacent shooting angles is recorded as the extended intersection point, and auxiliary lines perpendicular to the outer boundary line and the inner boundary line are drawn through the extended intersection point to obtain A segmentation lines; Using A segmentation lines as segmentation criteria, the overhead image of the sports field is divided into A adjacent sub-images, and the sports fields corresponding to the A sub-images are recorded as sub-fields.

3. A speed skating motion trajectory tracking and video acquisition method according to claim 2, characterized in that: The method for extracting the center offset value is: Count the motion durations of the A sub-trajectory videos one by one to obtain A motion durations, and record the sub-trajectory video corresponding to the minimum motion duration as the target video; Mark out at preset time intervals At a certain time point, the target video is captured. The image at each time point is obtained A moment image; After dividing the remaining A-1 exercise durations by the minimum value of the exercise duration, A-1 interval multiples are obtained, and the A-1 interval multiples are multiplied by the preset duration interval to obtain A-1 revised duration intervals; Take the A-1 revised time intervals as the standard and mark the remaining A-1 sub-track videos. At a certain time point, A-1 sub-trajectory videos are captured. An image at a certain point in time; exist The positions of the two feet of the trainee are respectively identified in the image at each moment, two foot regions are obtained, and all the pixels in the two foot regions are marked; The distance values ​​from the pixel points in one of the foot regions to the pixel points in the other foot region are measured one by one by using a scale, and the two pixel points corresponding to the minimum value of the distance value are recorded as the first pixel point and the second pixel point; The first pixel point and the second pixel point are connected to form a pixel line, and the minimum distance between the midpoint of the pixel line and the preset center line is measured to obtain Individual deviation value; Will The sub-deviation values ​​are accumulated and averaged to obtain A center offset values; The expression for the center offset value is: ； In the formula, For the The center offset value of the sub-track video, For the The sub-track video The sub-deviation value.

4. A speed skating motion trajectory tracking and video acquisition method according to claim 3, characterized in that: The method for extracting the average speed of movement is: In the A sub-track videos, the speed sensor detects the trainee’s The instantaneous speed at a certain time point is obtained Speed ​​of individual; After removing the maximum and minimum sub-speeds, the remaining The speeds of each individual are accumulated and averaged to obtain A average speed values ​​of the movements; The expression of the average speed of motion is: ； In the formula, For the The average speed of the sub-trajectory video, For the The sub-track video Speed ​​of the child.

5. A speed skating motion trajectory tracking and video acquisition method according to claim 4, characterized in that: The method for obtaining the center of gravity sway value is: exist In each moment image, mark the position of the trainee's head one by one, obtain the head area, and mark all the pixels in the head area; The distance values ​​from all the pixels in the head area to the ground of the sub-site are measured one by one using a scale to obtain The center of gravity distance values ​​are obtained by subtracting the maximum value of the center of gravity distance value from the minimum value of the center of gravity distance value to obtain A center of gravity sway values; The expression of the center of gravity sway value is: ； In the formula, For the The center of gravity shake value of each sub-track video, For the The maximum value of the center of gravity distance of the sub-trajectory videos, For the The minimum value of the center-of-gravity distance of the sub-trajectory videos; The method for generating the annotated sub-video is: The center offset value, average speed value and center of gravity shaking value of A sub-track videos are respectively compared with the offset safety value , mean safety value and sway safety value Compare; when Greater than When The center offset value of each sub-track video is recorded as the annotation data; when Less than When The average speed of each sub-trajectory video is recorded as the remark data; when Greater than When The center of gravity shaking value of each sub-trajectory video is recorded as the remark data; Remark boxes are constructed on the A sub-track videos respectively, and the remark data are imported into the remark boxes of the corresponding sub-track videos one by one to generate A remark sub-videos.

6. A motion trajectory tracking and video acquisition device, comprising a video controller, in which a site division module, an area planning module, a mode determination module, a video annotation module and a video splicing module are integrated, characterized in that: The site division module, the area planning module, the mode determination module, the video annotation module and the video splicing module are used together to implement a speed skating motion trajectory tracking and video acquisition method according to any one of claims 1 to 5, wherein: A venue division module is used to filter out the overhead image of the sports venue from the database, and divide the sports venue into adjacent sub-venues based on the venue division criteria; The regional planning module is used to obtain the comprehensive site data of the sub-site, including the critical curvature, linear area ratio and boundary span distance, calculate the site buffer coefficient, and plan the critical area in the sub-site by combining the site buffer coefficient and the standard area amplitude; A mode determination module is used to obtain the real-time position of the training personnel in the sub-venue and determine whether to execute the tracking and video acquisition switching mode based on the area switching criteria; The video annotation module is used to obtain the sub-track video of the trainee, extract the comprehensive motion data from the sub-track video, the comprehensive motion data includes the center offset value, the average speed value and the center of gravity sway value, and annotate the sub-track video to generate an annotated sub-video; The video splicing module is used to sequentially number the annotated sub-videos and splice them head to tail to generate a speed roller skating motion trajectory video of the training personnel.

Images