Stadium trajectory optimization method and device, wearable smart device, computer program product

Abstract

The application relates to the technical field of trajectory optimization, and further relates to a sports field trajectory optimization method and device, a wearable smart device and a computer program product. The method comprises the following steps: performing thinning processing on acquired sports field trajectory data to obtain target sample data; determining a first vector according to a first curve point and a first straight line point of the target sample data; determining a second vector according to the curvatures, position information and speeds of all trajectory points of the target sample data; determining a curvature threshold value according to the first vector and the second vector; identifying a second curve point and a second straight line point in all trajectory points according to the curvature threshold value and the curvatures; and fitting the second curve point and the second straight line point to determine a sports field trajectory. The first curve point and the first straight line point are obtained by marking all trajectory points by a user. The method not only makes the optimized sports field trajectory effectively fit the actual sports field shape, but also improves the use experience of the user.

Description

Technical Field

[0001] This application relates to the field of trajectory optimization technology, and further to a method and apparatus for optimizing sports field trajectories, wearable smart devices, and computer program products. Background Technology

[0002] In wearable smart devices, trajectory recording based on satellite positioning technology has been widely used. However, existing technologies commonly suffer from frequent satellite signal drift in relatively enclosed or semi-enclosed sports fields such as playgrounds. Therefore, when existing wearable smart devices present the trajectory by connecting the original trajectory points, they cannot effectively handle the data noise caused by drift, resulting in a chaotic, distorted trajectory graphic that cannot match the geometry of the actual sports field (such as a standard elliptical track).

[0003] Therefore, there is an urgent need for a trajectory optimization processing scheme for sports fields to generate smooth trajectories that fit the actual shape of the sports field. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a method and apparatus for optimizing sports field trajectories, a wearable smart device, and a computer program product, which significantly improves the user experience.

[0005] In a first aspect, this application provides a method for optimizing sports field trajectories, applied to wearable smart devices, comprising: thinning the acquired sports field trajectory data to obtain target sample data; determining a first vector based on the first curve point and the first straight point of the target sample data; determining a second vector based on the curvature, position information, and velocity of all trajectory points in the target sample data; determining a curvature threshold based on the first vector and the second vector; identifying a second curve point and a second straight point among all trajectory points based on the curvature threshold and curvature; fitting the second curve point and the second straight point to determine the sports field trajectory; wherein the first curve point and the first straight point are obtained by the user annotating all trajectory points.

[0006] The above method for optimizing sports field trajectories first thins the acquired sports field trajectory data to obtain target sample data. Thinning effectively reduces computational load to accommodate the resource constraints of wearable smart devices. Then, a first vector is determined based on the user-annotated first curve and first straight point. Simultaneously, a second vector is determined by combining the curvature, position information, and velocity of all trajectory points in the target sample data. Next, a curvature threshold is determined based on the first and second vectors, and the second curve and second straight point are identified using this threshold. Finally, the optimized sports field trajectory is generated by fitting the second curve and second straight point together. The optimized trajectory effectively conforms to the actual shape of the sports field, improving the user experience.

[0007] In one implementation, fitting the second curve point and the second straight point to determine the track trajectory specifically includes: determining the first curve segment and the second curve segment based on the least squares method and the second curve point; determining the first straight segment and the second straight segment based on the least squares method and the second straight point; and determining the track trajectory based on the first curve segment, the second curve segment, the first straight segment, and the second straight segment.

[0008] In one implementation, the algorithm used for thinning includes the Douglas-Puk thinning algorithm.

[0009] The above sports field trajectory optimization method uses the Douglas-Puk thinning algorithm to thin the acquired sports field trajectory data, which can remove redundant data while retaining key trajectory points. This not only makes the final optimized sports field trajectory more accurate, but also saves memory and speeds up system processing.

[0010] In one implementation, the acquired sports field trajectory data is thinned to obtain target sample data. Specifically, this includes: determining a target straight line based on the first and last trajectory points of the sports field trajectory data; determining the target distance from all trajectory points of the sports field trajectory data to the target straight line based on the target straight line and all trajectory points of the sports field trajectory data; determining a comprehensive feature score for each trajectory point of the sports field trajectory data based on the target distance, curvature, and rate of change of velocity corresponding to all trajectory points of the sports field trajectory data; and thinning all trajectory points of the sports field trajectory data based on the comprehensive feature score threshold and the comprehensive feature score, and using the thinned trajectory points as the target sample data.

[0011] In one implementation, the trajectory points of the sports field trajectory data are thinned out based on the comprehensive feature score threshold and the comprehensive feature score. Specifically, this includes: determining the maximum value of the comprehensive feature score based on the comprehensive feature score of each trajectory point in the sports field trajectory data; when the maximum value of the comprehensive feature score is not greater than the comprehensive feature score threshold, retaining the first and last trajectory points; when the maximum value of the comprehensive feature score is greater than the comprehensive feature score threshold, using the trajectory point corresponding to the maximum value of the comprehensive feature score as the midpoint, dividing all trajectory points in the sports field trajectory data into subsets, and repeating the above process on the subsets until all trajectory points have been processed.

[0012] The above method for optimizing sports field trajectories constructs a target straight line using the first and last trajectory points. It calculates the target distance from each trajectory point in the sports field trajectory data to the target straight line and integrates the target distance, curvature, and rate of change of velocity to generate a comprehensive feature score for each trajectory point. Then, based on a comprehensive feature score threshold, a recursive strategy is adopted: the trajectory subset is divided with the trajectory point corresponding to the maximum comprehensive feature score as the midpoint. If the maximum comprehensive feature score is not greater than the comprehensive feature score threshold, the first and last trajectory points are retained; otherwise, the same process is iterated on the subset until all data is processed. Compared to the Douglas-Puk thinning algorithm, which thins the acquired sports field trajectory data, this scheme improves the accuracy of the retained key trajectory points by integrating multiple feature data to determine the comprehensive feature score of each trajectory point. This makes the optimized sports field trajectory more closely resemble the actual shape of the sports field, while also removing redundant data, saving memory, and accelerating system processing speed.

[0013] In one implementation, the method further includes: using a filtering algorithm to filter the original trajectory data, and using the filtered original trajectory data as the motion field trajectory data.

[0014] In one implementation, a filtering algorithm is used to filter the original trajectory data, specifically including: setting a corresponding initial window for each trajectory point of the original trajectory data and calculating the local standard deviation of the trajectory points within the initial window; adjusting the corresponding initial window according to each local standard deviation; and filtering the corresponding trajectory points of the original trajectory data according to each adjusted initial window.

[0015] Secondly, this application provides a sports field trajectory optimization device applied to a wearable smart device, comprising: a feature analysis module configured to: perform thinning processing on acquired sports field trajectory data to obtain target sample data; determine a first vector based on a first curve point and a first straight point of the target sample data; determine a second vector based on the curvature, position information, and velocity of all trajectory points in the target sample data; determine a curvature threshold based on the first vector and the second vector; and identify a second curve point and a second straight point among all trajectory points based on the curvature threshold and curvature; and a fitting module configured to fit the second curve point and the second straight point to determine the sports field trajectory; wherein the first curve point and the first straight point are obtained by the user annotating all trajectory points.

[0016] Thirdly, this application provides a wearable smart device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the motion field trajectory optimization method implemented above.

[0017] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described motion field trajectory optimization methods.

[0018] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described motion field trajectory optimization methods.

[0019] Compared with the prior art, the present invention has at least one of the following beneficial effects: 1. First, the acquired sports field trajectory data is thinned to obtain target sample data. Thinning effectively reduces the computational load to adapt to the resource constraints of wearable smart devices. Then, a first vector is determined based on the user-annotated first curve and first straight point. Simultaneously, a second vector is determined by combining the curvature, position information, and velocity of all trajectory points in the target sample data. A curvature threshold is then determined based on the first and second vectors, and the second curve and second straight point are identified using this threshold. Finally, an optimized sports field trajectory is generated by fitting the second curve and second straight points. The optimized trajectory effectively conforms to the actual shape of the sports field, improving the user experience.

[0020] 2. By using the Douglas-Puk thinning algorithm to thin the acquired sports field trajectory data, redundant data can be removed while retaining key trajectory points. This not only makes the final optimized sports field trajectory more accurate, but also saves memory and speeds up system processing.

[0021] 3. Construct a target straight line using the first and last trajectory points. Calculate the target distance from each trajectory point in the sports field trajectory data to the target straight line. Integrate the target distance, curvature, and rate of change of velocity to generate a comprehensive feature score for each trajectory point. Then, based on the comprehensive feature score threshold, employ a recursive strategy: divide the trajectory into subsets using the trajectory point corresponding to the maximum comprehensive feature score as the midpoint. If the maximum comprehensive feature score is not greater than the comprehensive feature score threshold, retain the first and last trajectory points; otherwise, iterate the same process on the subsets until all are processed. Compared to the Douglas-Puk thinning algorithm, which thins the acquired sports field trajectory data, this scheme improves the accuracy of the retained key trajectory points by integrating multiple feature data to determine the comprehensive feature score of each trajectory point. This makes the optimized sports field trajectory more closely match the actual shape of the sports field, while also removing redundant data, saving memory, and accelerating system processing speed. Attached Figure Description

[0022] The preferred embodiments will now be described in a clear and easy-to-understand manner, in conjunction with the accompanying drawings, to further explain the above-mentioned characteristics, technical features, advantages, and implementation methods of the present invention.

[0023] Figure 1 A flowchart of a sports field trajectory optimization method provided in an embodiment of this application is shown; Figure 2 This document illustrates a flowchart of a method for determining target sample data according to an embodiment of this application. Figure 3 This invention illustrates another flowchart for determining target sample data provided in an embodiment of this application; Figure 4 This document illustrates a flowchart of filtering raw trajectory data according to an embodiment of this application. Figure 5 This paper shows a structural block diagram of a sports field trajectory optimization device provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of a wearable smart device provided in an embodiment of this application is shown. Detailed Implementation

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0025] To keep the drawings concise, each figure only schematically shows the parts relevant to the invention, and these do not represent the actual structure of the product. Furthermore, to facilitate understanding, in some figures, only one of components with the same structure or function is schematically depicted, or only one is labeled. In this document, "one" not only means "only one," but can also mean "more than one."

[0026] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0027] In this document, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0028] Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] It should be noted that the above embodiments can be freely combined as needed. The above are merely preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0030] In wearable smart devices, trajectory recording based on satellite positioning technology has been widely used. However, existing technologies commonly suffer from frequent satellite signal drift in relatively enclosed or semi-enclosed sports fields such as playgrounds. This phenomenon is mainly caused by factors such as building obstruction, vegetation interference, and complex electromagnetic environments, resulting in random offsets, discontinuities, or abnormal fluctuations in the collected trajectory point data. Therefore, when existing wearable smart devices present the trajectory by connecting the original trajectory points, they cannot effectively handle the data noise caused by drift, resulting in a chaotic, distorted trajectory graphic that cannot match the geometry of the actual sports field (such as a standard elliptical running track).

[0031] Currently, trajectory optimization schemes are commonly used for vehicle trajectory optimization. For example, this involves filtering and thinning the latitude and longitude data of a vehicle during its journey, and then plotting the trajectory curve on a map based on the processed latitude and longitude data to obtain the actual vehicle route. However, there is still no corresponding processing method for trajectory optimization of sports fields using wearable smart devices. Therefore, there is an urgent need for a trajectory optimization processing scheme for sports fields to generate smooth trajectories that closely match the actual shape of the sports field.

[0032] The following explanation is based on the accompanying diagram: Reference Appendix Figure 1 The diagram illustrates a flowchart of a sports field trajectory optimization method provided in an embodiment of this application. Figure 1 As shown, it includes: S100 performs thinning processing on the acquired sports field trajectory data to obtain target sample data.

[0033] S110, determine the first vector based on the first curve point and the first straight point of the target sample data.

[0034] S120, determine the second vector based on the curvature, position information and velocity of all trajectory points in the target sample data.

[0035] S130, determine the curvature threshold based on the first vector and the second vector.

[0036] S140, based on the curvature threshold and curvature, identifies the second curve point and the second straight point among all trajectory points.

[0037] S150, fits the second curve point and the second straight point to determine the track trajectory. The first curve point and the first straight point are obtained by the user by labeling all trajectory points.

[0038] Sports field trajectory data can refer to the trajectory data of a standard elliptical running track, such as, but not limited to, the trajectory data of running tracks in school playgrounds and football fields. The sports field trajectory data can be filtered or unfiltered data. Furthermore, the source of the sports field trajectory data can be a module or device with positioning function in a wearable smart device, or it can be collected and sent to the wearable smart device by other devices with positioning function; this application does not limit the source of the data. The positioning function can be implemented using, but is not limited to, GPS positioning systems, BeiDou positioning systems, Galileo positioning systems, etc.

[0039] The acquired sports field trajectory data is essentially a set of multiple trajectory points. The acquired sports field trajectory data is thinned using the Douglas-Puk thinning algorithm or a multi-feature fusion thinning algorithm to obtain a new set of trajectory points, which is then used as the target sample data.

[0040] After obtaining the target sample data, the user can label the first curve point and the first straight point in the target sample data. The first curve point can be labeled as 1, and the first straight point can be labeled as 0. Based on all the first curve points and all the first straight points, a first vector (or straight and curve label vector Y) is generated. The curvature of each trajectory point in the target sample data is calculated using the three-point curvature method. Based on the curvature, position information (position information refers to the coordinates of the trajectory points), and velocity of all trajectory points in the target sample data, a second vector (or feature vector X) is generated. Using the second vector as input and the first vector as output, a decision model is trained. The purpose of training the decision model is to obtain a curvature threshold, which can be the 90th percentile of the curvature of the curve point. When the curvature of a trajectory point is less than the curvature threshold, the trajectory point is a straight point; otherwise, the trajectory point is a curve point. Therefore, based on the curvature threshold and the curvature of all trajectory points in the target sample data, the curves and straights in the target sample data can be re-divided, thus determining the second curve and second straight points among all trajectory points in the target sample data. By fitting all the second curve points and all the second straight points using the least squares method, the optimized motion field trajectory can be obtained.

[0041] This embodiment first performs thinning processing on the acquired sports field trajectory data to obtain target sample data. Thinning processing can effectively reduce the computational load to adapt to the resource constraints of wearable smart devices. Then, a first vector is determined based on the user-annotated first curve point and first straight point. Simultaneously, a second vector is determined by combining the curvature, position information, and velocity of all trajectory points in the target sample data. Furthermore, a curvature threshold is determined based on the first and second vectors, and the second curve point and second straight point are identified based on this curvature threshold. Finally, an optimized sports field trajectory is generated by fitting the second curve point and second straight point. The optimized sports field trajectory effectively conforms to the actual shape of the sports field, improving the user experience.

[0042] When the track data represents a full lap of a standard elliptical track, fitting the data to all second curve points and all second straight points using the least squares method yields two straight segments and two curve segments. When the track data represents half a lap of a standard elliptical track, fitting the data to all second curve points and all second straight points using the least squares method yields half of two straight segments and one curve segment. When the track data represents other types of track data for a standard elliptical track, the corresponding proportions of straight and curve segments can be obtained.

[0043] The following example uses the trajectory data of a complete lap of a standard elliptical running track to illustrate the specific fitting process: all points on the second straightaway can form a point set. The least squares method is used to fit the straight line. Thus, the first objective function can be obtained. The first objective function is: By using straight lines Take the partial derivatives of parameters a and b and set them to 0, then solve the system of equations to obtain the values ​​of a and b. Substitute the obtained a and b into the first objective function to obtain the trajectory equation of the first straight track segment, which is the first straight track segment itself. Furthermore, since the two straight tracks on the track are parallel, the second straight track segment can be determined.

[0044] All curve segments can form a set of points. Hedianji Point set For example, the least squares method is used to fit the circle equation. To obtain the second objective function - ,right Take the partial derivatives of R and R respectively and set them to 0. Solve the system of equations to obtain the center of the circle. , The values ​​of the center and radius R. , Substituting the radius R into the second objective function, we can obtain the trajectory equation for the first curve segment, which is the first curve segment itself. Similarly, for the point set... By performing the same process, the second curve segment can be obtained. After obtaining the first straight segment, the second straight segment, the first curve segment, and the second curve segment, these straight and curve segments are connected sequentially to obtain the optimized track trajectory. Through the above processing, the track points of complex scenarios can be analyzed and fitted, improving the user experience.

[0045] In some thinning methods, this application can perform thinning processing on the acquired sports field trajectory data using either unmodified or modified thinning algorithms. For example, the unmodified Douglas-Puk thinning algorithm can be used directly. Alternatively, an improved multi-feature fusion thinning algorithm can be used.

[0046] The Douglas-Puk thinning algorithm works as follows: Connect the first and last trajectory points in the sports field trajectory data with a straight line, then calculate the distance between all intermediate trajectory points and this straight line, and find the maximum distance value. Compare the maximum distance value with the thinning threshold: If the maximum distance value is less than the thinning threshold, all intermediate trajectory points in the sports field trajectory data are discarded; if the maximum distance value is not less than the thinning threshold, the trajectory point corresponding to the maximum distance value is used as the intermediate point, and combined with the first and last trajectory points, the sports field trajectory data is divided into two subsets. Then, the above comparison process is repeated for these two subsets until no further subsets can be divided.

[0047] This application uses the Douglas-Puk thinning algorithm to thin the acquired sports field trajectory data, which can remove redundant data while retaining key trajectory points. This not only makes the final optimized sports field trajectory more accurate, but also saves memory and speeds up system processing.

[0048] The improved multi-feature fusion thinning algorithm adopts a similar approach to the Douglas-Puk thinning algorithm; please refer to the appendix for details. Figure 2 The diagram shows a flowchart for determining target sample data. Figure 2 As shown, it includes: S200 determines the target straight line based on the first and last trajectory points of the sports field trajectory data.

[0049] S210, Based on the target straight line and all trajectory points of the sports field trajectory data, determine the target distance from all trajectory points of the sports field trajectory data to the target straight line.

[0050] S220, based on the target distance, curvature, and rate of change of velocity corresponding to all trajectory points in the sports field trajectory data, determine the comprehensive feature score of each trajectory point in the sports field trajectory data.

[0051] S230, based on the comprehensive feature scoring threshold and comprehensive feature score, thins out all trajectory points in the sports field trajectory data, and uses the thinned trajectory points as target sample data.

[0052] Continue to refer to the appendix Figure 3 Step S230 includes: S300 determines the maximum value of the comprehensive feature score based on the comprehensive feature score of each trajectory point in the sports field trajectory data.

[0053] S310: When the maximum value of the comprehensive feature score is not greater than the comprehensive feature score threshold, retain the first and last trajectory points.

[0054] S320, when the maximum value of the comprehensive feature score is greater than the comprehensive feature score threshold, take the trajectory point corresponding to the maximum value of the comprehensive feature score as the midpoint, divide all trajectory points of the sports field trajectory data into subsets, and repeat the above process on the divided subsets until all trajectory points have been processed.

[0055] The acquired sports field trajectory data is essentially a set of multiple trajectory points. Connecting the first and last trajectory points of this set yields the target straight line. The distances from all trajectory points in the sports field trajectory data to the target straight line are calculated, or the distances from all intermediate trajectory points in the sports field trajectory data to the target straight line are calculated (the distances from the first and last trajectory points to the target straight line are 0). The curvature of all trajectory points in the sports field trajectory data is calculated using the three-point curvature method, and the rate of change of velocity corresponding to each trajectory point is calculated based on the time intervals between each trajectory point. The target distance, curvature, and rate of change of velocity of all trajectory points in the sports field trajectory data are then substituted into the formula. This yields a comprehensive feature score for each trajectory point in the sports field trajectory data. Among these features, For comprehensive feature scoring, For the target distance, For curvature, For the rate of change of velocity, , , As weight, and + + =1.

[0056] The processing approach after obtaining the comprehensive feature score corresponding to each trajectory point is similar to the Douglas-Puk thinning algorithm. Based on the comprehensive feature score of each trajectory point in the sports field trajectory data, the maximum value of the comprehensive feature score is determined. The maximum value of the comprehensive feature score is compared with the comprehensive feature score threshold: if the maximum value of the comprehensive feature score is not greater than the comprehensive feature score threshold, all intermediate trajectory points in the sports field trajectory data are discarded (if the trajectory point corresponding to the maximum value of the comprehensive feature score is the first or last trajectory point, the first and last trajectory points are directly retained, and all intermediate trajectory points are discarded; if the trajectory point corresponding to the maximum value of the comprehensive feature score is an intermediate trajectory point, all intermediate trajectory points are discarded); if the maximum value of the comprehensive feature score is greater than the comprehensive feature score threshold, the trajectory point corresponding to the maximum value of the comprehensive feature score is used as the midpoint and combined with the first and last trajectory points to divide all trajectory points in the sports field trajectory data into two subsets. Then, the aforementioned comparison process is repeated for these two subsets until no further subsets can be divided.

[0057] Whether using the Douglas-Puk thinning algorithm or the improved multi-feature fusion thinning algorithm, after obtaining the target sample data, the aforementioned processes of determining the first vector, determining the second vector, determining the curvature threshold, determining the second curve point and the second straight point, and specifically fitting the second curve point and the second straight point to obtain the optimized motion field trajectory are required. The specific execution process is the same as in the aforementioned embodiments, and will not be repeated here.

[0058] This application's embodiment constructs a target straight line using the first and last trajectory points, calculates the target distance from each trajectory point in the sports field trajectory data to the target straight line, and integrates the target distance, curvature, and rate of change of velocity to generate a comprehensive feature score for each trajectory point. Then, based on a comprehensive feature score threshold, a recursive strategy is adopted: the trajectory subset is divided with the trajectory point corresponding to the maximum comprehensive feature score as the midpoint; if the maximum comprehensive feature score is not greater than the comprehensive feature score threshold, the first and last trajectory points are retained; otherwise, the same process is iteratively executed on the subset until all processing is complete. Compared to the Douglas-Puk thinning algorithm's method of thinning the acquired sports field trajectory data, this scheme, by integrating multiple feature data to determine the comprehensive feature score of each trajectory point, can further improve the accuracy of the retained key trajectory points, making the optimized sports field trajectory more closely match the actual sports field shape. Simultaneously, it can remove redundant data, save memory, and accelerate system processing speed.

[0059] As mentioned in the foregoing embodiments, the acquired sports field trajectory data can be either filtered or unfiltered. When the acquired sports field data is filtered, the filtering algorithm used in this application may include, but is not limited to, a 5-point smoothing filter algorithm. Filtering the original trajectory data using a filtering algorithm can effectively remove noise and improve the accuracy of the sports trajectory.

[0060] For details on the 5-point smoothing filter algorithm, please refer to the appendix. Figure 4 This illustrates a flowchart of filtering raw trajectory data according to an embodiment of this application. Figure 4 As shown, it includes: S400 sets a corresponding initial window for each trajectory point in the original trajectory data and calculates the local standard deviation of the trajectory points within the initial window.

[0061] S410 adjusts the corresponding initial window based on each local standard deviation.

[0062] S420 filters the corresponding trajectory points of the original trajectory data based on each adjusted initial window.

[0063] Let the coordinate sequence of the original trajectory data be Pi(xi,yi). For each trajectory point Pi, an initial window is set with Pi as the center. (The initial window size is set to 5). Then, the local standard deviation of the trajectory points within each initial window can be calculated using the following formula: ,in, = , = , For window The number of trajectory points within the range.

[0064] After obtaining all local standard deviations, set a threshold. and threshold Based on local standard deviation and threshold and threshold Adjust the initial window corresponding to each trajectory point in the original trajectory data. The specific adjustment method is as follows: = .in, Within the range of 3-9.

[0065] Based on each adjusted initial window, the corresponding trajectory point Pi of the original trajectory data is smoothed and filtered to obtain new trajectory points. Finally, all the new trajectory points This constitutes the motion field trajectory data obtained as mentioned in the aforementioned embodiments. The formula for determining the new trajectory point is: =

[0066] = .

[0067] Then, the data is thinned using the Douglas-Puk thinning algorithm described in the previous embodiment, or using an improved multi-feature fusion thinning algorithm. After obtaining the target sample data, the process of determining the first vector, determining the second vector, determining the curvature threshold, determining the second curve point and the second straight point, and specifically fitting the second curve point and the second straight point to obtain the optimized motion field trajectory is executed. The specific execution process is the same as in the previous embodiment, and will not be repeated here.

[0068] Reference Appendix Figure 5 The diagram illustrates a structural block diagram of a sports field trajectory optimization device provided in an embodiment of this application. Figure 5 As shown, the device 500 includes: a feature analysis module 510, configured to: perform thinning processing on the acquired sports field trajectory data to obtain target sample data; determine a first vector based on the first curve point and the first straight point of the target sample data; determine a second vector based on the curvature, position information, and velocity of all trajectory points in the target sample data; determine a curvature threshold based on the first vector and the second vector; and identify the second curve point and the second straight point among all trajectory points based on the curvature threshold and curvature; and a fitting module 520, configured to fit the second curve point and the second straight point to determine the sports field trajectory; the first curve point and the first straight point are obtained by the user annotating all trajectory points.

[0069] This embodiment first performs thinning processing on the acquired sports field trajectory data to obtain target sample data. Thinning processing can effectively reduce the computational load to adapt to the resource constraints of wearable smart devices. Then, a first vector is determined based on the user-annotated first curve point and first straight point. Simultaneously, a second vector is determined by combining the curvature, position information, and velocity of all trajectory points in the target sample data. Furthermore, a curvature threshold is determined based on the first and second vectors, and the second curve point and second straight point are identified based on this curvature threshold. Finally, an optimized sports field trajectory is generated by fitting the second curve point and second straight point. The optimized sports field trajectory effectively conforms to the actual shape of the sports field, improving the user experience.

[0070] Reference Appendix Figure 6This application also provides a wearable smart device 600, which includes a memory 610, a processor 620, and a computer program stored in the memory 610. The processor 620 executes the computer program to implement the steps of the motion field trajectory optimization method of any of the above embodiments.

[0071] The memory 610 can be non-volatile memory (NVM), such as, but not limited to, semiconductor non-volatile memory, disk storage, or optical storage. Semiconductor non-volatile memory includes, but is not limited to, read-only memory (ROM) or flash memory, such as mask ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), NAND flash memory, or NOR flash memory.

[0072] The memory 610 can also be volatile memory, such as random access memory (RAM). RAM includes, for example, static random-access memory (SRAM) or dynamic random-access memory (DRAM). DRAM includes, for example, synchronous dynamic RAM (SDRAM) or double data rate SDRAM (DDR). With the development of technology, DDR includes, but is not limited to, DDR1, DDR2, DDR3, ..., DDR5, and may also include future DDR6.

[0073] Processor 620 is a circuit with signal processing capabilities. In one example, the processor can be a circuit with instruction read and execute capabilities; such as a central processing unit (CPU), microcontroller unit (MCU), microprocessor unit (MPU), graphics processing unit (GPU), or digital signal processor (DSP). In another example, the processor can realize its processing capabilities through the logical relationships of hardware circuits, which can be fixed or reconfigurable; for example, the processor can be a dedicated processor, such as a processor implemented with an application-specific integrated circuit (ASIC), which realizes its processing capabilities through the design of the logical relationships between components within the circuit; or a processor implemented with a programmable logic device (PLD), which realizes its processing capabilities by configuring the logical relationships between logic devices through configuration files; for example, a processor implemented with a field-programmable gate array (FPGA). In another example, the processor can be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), tensor processing unit (TPU), deep learning processing unit (DPU), etc. This application is not limited to the type of processor.

[0074] In some embodiments of this application, wearable smart devices include, but are not limited to, smartwatches, smart bracelets, etc.

[0075] The wearable smart device used in this application embodiment is basically similar to the method embodiment, so the description is relatively simple. For relevant details, please refer to the description of the method embodiment.

[0076] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the motion field trajectory optimization method of any of the above embodiments.

[0077] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the motion field trajectory optimization method of any of the above embodiments.

[0078] It should be noted that the above embodiments can be freely combined as needed. The above are merely preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing sports field trajectories, applied to wearable smart devices, characterized in that, include: The acquired sports field trajectory data is thinned to obtain target sample data; Based on the first curve point and the first straight point of the target sample data, determine the first vector; The second vector is determined based on the curvature, position information, and velocity of all trajectory points in the target sample data; Determine the curvature threshold based on the first vector and the second vector; Based on the curvature threshold and the curvature, identify the second curve point and the second straight point among all trajectory points; Fit the second curve point and the second straight point to determine the trajectory of the motion field; The first curve point and the first straight point are obtained by the user by marking all trajectory points.

2. The method for optimizing sports field trajectories according to claim 1, characterized in that, The process of fitting the second curve point and the second straight point to determine the trajectory of the sports field specifically includes: Based on the least squares method and the second curve point, determine the first curve segment and the second curve segment; Based on the least squares method and the second straight point, determine the first straight segment and the second straight segment; The trajectory of the sports field is determined based on the first curved section, the second curved section, the first straight section, and the second straight section.

3. The method for optimizing sports field trajectories according to claim 1, characterized in that, The algorithm used for thinning includes the Douglas-Puk thinning algorithm.

4. The method for optimizing sports field trajectories according to claim 1, characterized in that, The process of thinning the acquired sports field trajectory data to obtain target sample data specifically includes: The target straight line is determined based on the first and last trajectory points of the sports field trajectory data; Based on the target straight line and all trajectory points of the sports field trajectory data, determine the target distance from all trajectory points of the sports field trajectory data to the target straight line; Based on the target distance, curvature, and rate of change of velocity corresponding to all trajectory points in the sports field trajectory data, a comprehensive feature score is determined for each trajectory point in the sports field trajectory data. Based on the comprehensive feature scoring threshold and the comprehensive feature score, all trajectory points of the sports field trajectory data are thinned out, and the thinned trajectory points are used as the target sample data.

5. The method for optimizing sports field trajectories according to claim 4, characterized in that, The step of thinning all trajectory points in the sports field trajectory data based on the comprehensive feature scoring threshold and the comprehensive feature score specifically includes: Based on the comprehensive feature score of each trajectory point in the sports field trajectory data, determine the maximum value of the comprehensive feature score; When the maximum value of the comprehensive feature score is not greater than the comprehensive feature score threshold, the first and last trajectory points are retained. When the maximum value of the comprehensive feature score is greater than the comprehensive feature score threshold, the trajectory point corresponding to the maximum value of the comprehensive feature score is used as the midpoint to divide all trajectory points of the sports field trajectory data into subsets, and the aforementioned process is repeated for the divided subsets until all trajectory points have been processed.

6. The method for optimizing sports field trajectories according to claim 1, characterized in that, Also includes: The original trajectory data is filtered using a filtering algorithm, and the filtered original trajectory data is used as the trajectory data of the sports field.

7. The method for optimizing sports field trajectories according to claim 6, characterized in that, The filtering algorithm used to filter the original trajectory data specifically includes: Set a corresponding initial window for each trajectory point of the original trajectory data, and calculate the local standard deviation of the trajectory points within the initial window; Adjust the corresponding initial window based on each of the local standard deviations; Based on each adjusted initial window, the corresponding trajectory points of the original trajectory data are filtered.

8. A sports field trajectory optimization device, applied to wearable smart devices, characterized in that, include: The feature analysis module is configured to: perform thinning processing on the acquired motion field trajectory data to obtain target sample data; Based on the first curve point and the first straight point of the target sample data, determine the first vector; Based on the curvature, position information, and velocity of all trajectory points in the target sample data, a second vector is determined; based on the first vector and the second vector, a curvature threshold is determined. Based on the curvature threshold and the curvature, identify the second curve point and the second straight point among all trajectory points; The fitting module is configured to fit the second curve point and the second straight point to determine the trajectory of the track; the first curve point and the first straight point are obtained by the user by labeling all trajectory points.

9. A wearable smart device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the sports field trajectory optimization method according to any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the sports field trajectory optimization method according to any one of claims 1-7.

Images