A method and device for detecting the landing point area of ​​a space moving object

By obtaining and processing the coordinates of the object to be measured in continuous video frames, and combining coordinate conversion rules, accurately determine the landing area of ​​the moving object, solving the problem of inaccurate detection of the landing area in the prior art, improving the accuracy of the score and efficiency.

CN114494949BActive Publication Date: 2024-11-19SHENZHEN FAIRPLAY SPORTS DEV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210021881.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-10
Publication Date
2024-11-19
Estimated Expiration
2042-01-10

AI Technical Summary

Technical Problem

In the sports industry, it is difficult for the prior art to accurately detect the landing area of ​​space moving objects (such as balls), resulting in inaccurate ratings.

Method used

By obtaining the continuous video frames of the object to be tested, determining its pending coordinates under the image coordinate system, and then determining the target coordinates and target video frames. Combining the coordinate conversion rules, the range coordinates and target coordinates under the image coordinate system are converted into a top view score The plane coordinate system of the area, thereby determining the target area of ​​the landing point.

Benefits of technology

The accuracy and judgment efficiency of the landing site area are improved, ensuring the fairness and accuracy of the score.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114494949B_ABST
    Figure CN114494949B_ABST
Patent Text Reader

Abstract

The present application provides a method, device, equipment and storage medium for detecting the landing point area of ​​a moving object in space, the method comprising: obtaining continuous video frames of the object to be detected, determining the first undetermined coordinates of the object to be detected according to the video frames; determining the first target coordinates of the object to be detected according to the first undetermined coordinates, and determining the corresponding target video frame according to the first target coordinates; according to the coordinate conversion rule, converting the first range coordinates and the first target coordinates on the target video frame into the corresponding second range coordinates and second target coordinates in the plane coordinate system established based on looking down at the scoring area; determining the target area of ​​the landing point according to the second range coordinates and the second target coordinates. This technical solution can achieve accurate acquisition of the target video frame of the landing point, and improves the accuracy of judging the target area of ​​the landing point.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of image detection technology, and in particular to a method and device for detecting a landing point area of ​​a space moving object. Background Art

[0002] With the development of artificial intelligence, image and video processing is becoming more and more important. Automation will become the mainstream in the future. At present, the application of this aspect in the sports industry is becoming more and more common. On the one hand, it reduces labor costs, and on the other hand, it ensures the concept of absolute fairness, justice and openness. For example, in the tests, competitions and examinations in the sports industry, the detection equipment uses image and video processing to intelligently determine the landing point of the ball, and then determine the landing area of ​​the ball for scoring. However, due to the fast speed of the ball, the landing frame may be missed in a short time, which may lead to misjudgment of the area. At this time, it is particularly important to accurately judge the landing point area. Summary of the invention

[0003] The present application provides a method, device, equipment and storage medium for detecting a landing point area of ​​a space moving object, so as to solve the technical problem of inaccurate detection of the landing area of ​​a moving object.

[0004] In a first aspect, a method for detecting a landing point area of ​​a moving object in space is provided, which is applied to a terminal device or a detection device capable of detecting a landing point, the method comprising: acquiring continuous video frames of the object to be detected, determining first undetermined coordinates of the object to be detected according to the video frames, wherein the first undetermined coordinates are the position coordinates of the object to be detected in each video frame in an image coordinate system; determining first target coordinates of the object to be detected according to the first undetermined coordinates, and determining a corresponding target video frame according to the first target coordinates, wherein the first target coordinates are the position coordinates of the landing point of the object to be detected in the image coordinate system, and the target video frame is a video frame in which the landing point of the object to be detected is detected; according to a coordinate conversion rule, converting first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on overlooking the scoring area, wherein the first range coordinates are the coordinates of a point in the first scoring area determined based on the image coordinate system; determining a target area of ​​the landing point according to the second range coordinates and the second target coordinates, wherein the target area is the scoring area into which the object to be detected falls.

[0005] This technical solution determines the first target coordinates of the object to be measured by the first undetermined coordinates of the object to be measured in the continuous video frames, so as to accurately obtain the target video frame of the landing point. In addition, according to the coordinate conversion rule, the first range coordinates and the first target coordinates on the target video frame are converted into the corresponding second range coordinates and the second target coordinates, so that the target area of ​​the landing point can be intuitively determined based on the second range coordinates and the second target coordinates based on the plane overlooking the landing area, thereby improving the accuracy of judging the target area of ​​the landing point.

[0006] In combination with the first aspect, in a possible implementation, determining the first target coordinates of the object to be measured based on the first undetermined coordinates includes: using the first undetermined coordinates of the object to be measured in continuous video frames to determine the motion trajectory of the object to be measured during movement, and determining the predicted coordinate axis corresponding to the landing direction of the object to be measured in the image coordinate system; confirming the landing point based on the coordinate value of the first undetermined coordinate on the predicted coordinate axis, and determining the first target coordinates. The first undetermined coordinates have multiple coordinate values, and the predicted coordinate axis corresponding to the landing direction of the object to be measured is obtained through the motion trajectory of the object to be measured, and only the coordinate points on the predicted coordinate axis are used to confirm the landing point, which simplifies the determination conditions of the landing point and improves the determination efficiency.

[0007] In combination with the first aspect, in a possible implementation, the first target coordinates are determined according to the coordinate points on the predicted coordinate axis corresponding to the first undetermined coordinates, including: obtaining the coordinate value under the predicted coordinate axis corresponding to each of the first undetermined coordinates, obtaining the maximum coordinate value or the minimum coordinate value among the coordinate values ​​under the predicted coordinate axis; obtaining the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value, and using the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value as the first target coordinate. The motion trajectory of the object to be measured during movement can be determined by the changes in multiple first undetermined coordinates, and the first target coordinates are determined by obtaining the maximum coordinate value or the minimum coordinate value from the coordinate values ​​under multiple predicted coordinate axes, thereby effectively improving the accuracy of determining the first target coordinates.

[0008] In combination with the first aspect, in a possible implementation, the first range coordinates and the first target coordinates on the target video frame are converted into corresponding second range coordinates and second target coordinates in a plane coordinate system established by looking down on the scoring area according to a coordinate conversion rule, including: obtaining a coordinate conversion matrix of the preset image coordinate system and a plane coordinate system established by looking down on the scoring area; using the coordinate conversion matrix to convert the first range coordinates and the first target coordinates into corresponding second range coordinates and the second target coordinates. By converting the first range coordinates and the first target coordinates into the second range coordinates and the second target coordinates of the plane coordinate system established by looking down on the scoring area through the coordinate conversion matrix, the landing point area of ​​the object to be measured can be judged more effectively and intuitively.

[0009] In combination with the first aspect, in a possible implementation method, determining the target area of ​​the landing point according to the second range coordinates and the second target coordinates includes: determining the second scoring area in a plane coordinate system established based on the overlooking of the scoring area according to the second range coordinates; determining the target area of ​​the landing point according to the position of the second target coordinates in the second scoring area.

[0010] In combination with the first aspect, in a possible implementation manner, determining the corresponding target video frame according to the first target coordinates includes: determining the video frame to which the first target coordinates belong, and using the video frame to which the first target coordinates belong as the target video frame.

[0011] In combination with the first aspect, in a possible implementation method, determining the first undetermined coordinates of the object to be detected based on the video frame includes: detecting the first undetermined coordinates of the object to be detected in the video frame through a convolutional neural network target detection algorithm.

[0012] In a second aspect, a landing point area detection device for a space moving object includes:

[0013] A module for acquiring coordinates to be determined, used for acquiring continuous video frames of the object to be measured, and determining first coordinates to be determined of the object to be measured according to the video frames;

[0014] A target coordinate determination module, used to determine the first target coordinates of the object to be measured according to the first undetermined coordinates, and determine the corresponding target video frame according to the first target coordinates;

[0015] A coordinate conversion module, configured to convert the first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on looking down at the scoring area according to a coordinate conversion rule;

[0016] A target area determination module is used to determine the target area of ​​the landing point according to the second range coordinates and the second target coordinates.

[0017] In a third aspect, a computer device is provided, comprising a memory and one or more processors, wherein the one or more processors are used to execute one or more computer programs stored in the memory, and when the one or more processors execute the one or more computer programs, the computer device implements the landing point area detection method of a moving object in space according to the first aspect.

[0018] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the landing point area detection method of a space moving object according to the first aspect.

[0019] The present application can achieve the following technical effects: by determining the first target coordinates of the object to be measured by the first undetermined coordinates of the object to be measured in the continuous video frames, the target video frame of the landing point can be accurately obtained. In addition, according to the coordinate conversion rule, the first range coordinates and the first target coordinates on the target video frame are converted into the corresponding second range coordinates and the second target coordinates, so that the target area of ​​the landing point can be intuitively determined based on the second range coordinates and the second target coordinates based on the plane of the landing area, thereby improving the accuracy of judging the target area of ​​the landing point. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a flow chart of a method for detecting a landing point area of ​​a space moving object provided in an embodiment of the present application;

[0021] Figure 2 A schematic diagram of the network structure of a convolutional neural network target detection algorithm provided in an embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of a landing point area detection device for a space moving object provided in an embodiment of the present application;

[0023] Figure 4 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solution in the embodiment of the present application will be described below in conjunction with the accompanying drawings in the embodiment of the present application. The technical solution of the present application can be applied to a terminal device with a video image detection function, such as an image detector, etc. In this embodiment, the terminal device includes a video frame processor and a camera device, wherein the video frame processor is used to obtain a video frame, and the camera device is used to capture an image.

[0025] See also Figure 1 , Figure 1 A flowchart of a method for detecting a landing point area of ​​a space moving object provided in an embodiment of the present application is shown as follows: Figure 1 As shown, the method comprises the following steps:

[0026] S101, obtaining continuous video frames of the object to be measured, and determining the first undetermined coordinates of the object to be measured according to the video frames. The first undetermined coordinates are the position coordinates of the object to be measured in each video frame in an image coordinate system, and the image coordinate system is a coordinate system established based on a two-dimensional photo taken by a camera, and is used to specify the position of the object in the photo.

[0027] Since the present application is to determine the landing area of ​​a moving object in an image, before obtaining continuous video frames of the object to be tested, the moving process of the object to be tested is recorded in advance, and each frame of the video stream is obtained from the recorded video to obtain a video frame, wherein a video frame can be understood as an image captured by the camera at a certain moment. Since the object to be tested may be moving at a relatively fast speed, a general camera will produce a virtual image when capturing an image, so this embodiment can use a high frame rate camera for image capture. In a feasible implementation method, a Hikvision DS-2PT7T20IW-DE that supports a maximum high-definition screen output of 1920×1080@60fps can be used to improve the high definition of the image, thereby improving the accuracy of image capture.

[0028] Since the video frame is an image of the object to be measured captured by the camera at a certain moment, the position of the object to be measured at continuous moments during the movement process can be obtained according to the video frame, that is, the first undetermined coordinates can be obtained. It can be seen that the corresponding multiple first undetermined coordinates in multiple continuous video frames can reflect the motion trajectory of the object to be measured during the movement process. Therefore, the shorter the time interval between each acquired video frame, the smaller the change value between the acquired first undetermined coordinates, and thus the more accurate the landing point position coordinates of the object to be measured determined later according to the first undetermined coordinates. In order to obtain more accurate first undetermined coordinates through video frames, in this embodiment, when the object to be measured is moving at high speed, the processing speed of the video frame processor can reach 60 frames per second.

[0029] In this embodiment, the first undetermined coordinate can be detected by a deep learning neural network detection algorithm. Specifically, the deep learning neural network detection algorithm can be a convolutional neural network target detection algorithm, see Figure 2 , Figure 2It is the network structure of the convolutional neural network target detection algorithm, which consists of three parts: BackBone, PANet and Output. Among them, BackBone is the backbone network, which is used for feature extraction and represents a part of the network. It is generally used for front-end extraction of image information and generation of feature maps for use by the subsequent network; PANet is the path aggregation network, which aims to promote the flow of information; Output is the output segment. Bottle neck CSP is the bottleneck layer, and Conv is the convolution layer. SPP is used to make the input image size unrestricted. UpSample is upsampling, which is used to insert new elements between pixels based on the original image pixels using a suitable interpolation algorithm. Concat is an overlay layer, which is an overlay in the horizontal or vertical space. When using the convolutional neural network target detection algorithm, choose BackBone to use CSPDarknet. This network has the advantages of fewer parameters and deep convolution layers, which can improve detection accuracy. In the neck segment, PANet is used, and the feature pyramid method is used to expand the receptive field of the feature layer, so that the spatial information and semantic information of the object to be tested are combined to better detect the position of the object to be tested in the image field of view. In the output segment, three feature maps of different sizes take into account the detection of targets with different resolutions. In this embodiment, the above-mentioned convolutional neural network target detection algorithm can be used to detect the object to be tested in each frame of the continuous video frames to determine the first undetermined coordinates of the object to be tested in each video frame.

[0030] After obtaining the first undetermined coordinates of the object to be measured, execute Figure 1 In step S102, the first target coordinates of the object to be measured can be determined according to the first undetermined coordinates of the object to be measured in each video frame, and the corresponding target video frame can be determined according to the first target coordinates. The first target coordinates are the position coordinates of the landing point of the object to be measured in the image coordinate system, and the target video frame is the video frame in which the landing point of the object to be measured is detected.

[0031] When the object to be measured moves and falls or rebounds and rises in space, the first undetermined coordinate will have a specific change rule. According to this rule, the coordinate axis for determining the landing point of the object to be measured can be selected, and the coordinates of the landing point can be determined based on the coordinate values ​​on the coordinate axis. In this embodiment, the first undetermined coordinates of the object to be measured in the continuous video frames are used to determine the motion trajectory of the object to be measured during the movement, and the landing direction of the object to be measured in the image coordinate system is determined to correspond to the predicted coordinate axis. For example, when the X-axis of the image coordinate system is the horizontal direction, with the right as positive, and the Y-axis of the image coordinate system is the vertical direction, with the bottom as positive, it is obvious that when the object to be measured falls in space, the Y-axis of the first undetermined coordinate will continue to increase, and when the object to be measured bounces up in space, the Y-axis of the first undetermined coordinate will continue to decrease. Therefore, the object to be measured will touch the ground and bounce when it falls to the lowest point. According to this rule, the Y-axis can be selected as the predicted coordinate axis, and the Y-axis coordinate of the object to be measured when it contacts the ground can be determined based on the change of the coordinate value of the Y-axis, so that the landing coordinates of the object to be measured can be determined. Of course, the coordinate direction setting of the image coordinate system is not limited to the above setting. Finally, the landing point is confirmed according to the coordinate value of the first undetermined coordinate on the predicted coordinate axis to determine the first target coordinate.

[0032] Specifically, in this embodiment, the maximum coordinate value or the minimum coordinate value in the coordinate value under the predicted coordinate axis is selected, and the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value is obtained, and the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value is used as the first target coordinate. For example, in a volleyball test, based on the coordinate setting in the above embodiment, an image coordinate system is established with the upper left corner of the half side of the volleyball court as the origin (0,0), and 9 video frames are captured from the video. From the first frame to the ninth frame, the first undetermined coordinates of the volleyball in the corresponding video frames are Q1 (1035, 390), Q2 (1050, 401), Q3 (1067, 412), Q4 (1082, 423), Q5 (1095, 430), Q 6 (1104, 419), Q 7 (1110, 406), Q8 (1118, 394), and Q9 (1124, 383). It can be seen that the horizontal coordinate is increasing, the vertical coordinate is increasing in the first 5 frames, and decreasing in the last 4 frames. It can be analyzed that the first 5 frames are the process of the volleyball falling, and the last 4 frames are the process of the volleyball bouncing after landing. At this time, it can be judged that the first undetermined coordinate (1095, 430) corresponding to the Y-axis coordinate value of 430 is the landing point coordinate, that is, the first target coordinate of the volleyball, that is, the first undetermined coordinate corresponding to the maximum coordinate value in the coordinate value under the predicted coordinate axis is used as the first target coordinate. Finally, the video frame to which the first target coordinate belongs is determined according to the first target coordinate, and the video frame to which the first target coordinate belongs is the target video frame, so it can be judged that the ball is the target video frame in the 5th frame. In another possible implementation method, when the coordinate system is set to be positive in the Y-axis upward, it can be obtained according to the above logic that the first undetermined coordinate corresponding to the minimum coordinate value of the Y-axis is the first target coordinate, which will not be described in detail here.

[0033] The motion trajectory of the object to be measured during the movement process can be determined by the changes in multiple first pending coordinates. The first target coordinates are determined by obtaining the maximum coordinate value or the minimum coordinate value from the coordinate values ​​under multiple predicted coordinate axes, which effectively improves the accuracy of determining the first target coordinates.

[0034] After the target video frame is acquired, step S103 is executed to convert the first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on looking down at the scoring area according to the coordinate conversion rule. The first range coordinates are the coordinates of the points determining the first scoring area based on the image coordinate system.

[0035] The final implementation of the present application scheme is to determine the scoring area where the object to be tested falls. Therefore, before all steps are executed, the scoring area must be divided first. In this embodiment, the scoring area can be determined according to the key point coordinates of the scoring area, that is, the first range coordinates. For example, the scoring area of ​​the volleyball court can be determined according to the coordinates of the two end points of each straight line dividing the area as the first range coordinates, and the scoring area of ​​the table tennis court can be determined according to the coordinates of the four intersection points of the area boundary line as the first range coordinates. After the first range coordinates are determined, the first range coordinates are calibrated in the image coordinate system on the video frame to determine the first scoring area in the image. Since the first scoring area does not present a standard, real geometric shape in the image, in order to be able to intuitively and accurately determine the scoring area where the first target coordinates are located, it is necessary to convert the first range coordinates and the first target coordinates to the second range coordinates and the second target coordinates of the plane coordinate system established based on the overlooking of the scoring area. In this embodiment, the coordinate conversion is performed according to the coordinate conversion relationship between the image coordinate system and the plane coordinate system established based on the overlooking of the scoring area. The coordinate conversion relationship can be a conversion matrix, that is, the first range coordinates and the first target coordinates are converted into corresponding second range coordinates and second target coordinates using the coordinate conversion matrix. By converting the first range coordinates and the first target coordinates into the second range coordinates and the second target coordinates of a plane coordinate system established based on overlooking the scoring area through a coordinate conversion matrix, the landing point area of ​​the object to be measured can be determined more effectively and intuitively.

[0036] After obtaining the second range coordinates and the second target coordinates, step S104 may be continued to determine the target area of ​​the landing point according to the second range coordinates and the second target coordinates, wherein the target area is the scoring area where the object to be measured falls.

[0037] By calibrating the second range coordinates on the plane coordinate system established based on the overhead view of the scoring area, the real geometric shape of the scoring area can be restored according to the second range coordinates, that is, the second scoring area is obtained. At the same time, by calibrating the second target coordinates on the plane coordinate system established based on the overhead view of the scoring area, the second target coordinate point, that is, the landing point of the object to be measured, can be directly obtained. According to the positional relationship between the landing point and the second scoring area, the target area of ​​the object to be measured can be determined. In a feasible method, the target area where the landing point of the object to be measured falls can be directly determined according to which end of the regional boundary line formed by the second range coordinates the second target coordinates fall. For example, the plane coordinate system is established with the lower left corner as the origin, the X-axis is the horizontal direction, rightward is positive, and the Y-axis is the vertical direction, upward is positive. When the scoring area is a right triangle, based on the plane coordinate system, when the second range coordinates are (0, 0), (0, 100), and (100, 40), three boundary line equations can be determined according to the three second range coordinates. The area enclosed by the three boundary lines is the second scoring area. According to the second target coordinates and the three boundary line equations, it is determined at which end of the three boundary line equations the second target area is, so as to determine whether it falls into this triangular scoring area.

[0038] The method of the present application is introduced above. In order to better implement the method of the present application, the device of the present application is introduced next.

[0039] See also Figure 3 , Figure 3 Schematic diagram of a device for detecting a landing point area of ​​a space moving object provided in an embodiment of the present application. Figure 3 As shown, the landing point area detection device 30 of the space moving object includes:

[0040] The undetermined coordinates acquisition module 301 is used to acquire continuous video frames of the object to be measured, and determine the first undetermined coordinates of the object to be measured according to the video frames;

[0041] A target coordinate determination module 302 is used to determine the first target coordinates of the object to be measured according to the first undetermined coordinates, and determine the corresponding target video frame according to the first target coordinates;

[0042] A coordinate conversion module 303 is used to convert the first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established by looking down at the scoring area according to a coordinate conversion rule;

[0043] The target area determination module 304 is used to determine the target area of ​​the landing point according to the second range coordinates and the second target coordinates.

[0044] In a possible design, the target coordinate determination module 301 is specifically used to detect the first undetermined coordinates of the object to be detected in the video frame through a convolutional neural network target detection algorithm.

[0045] In one possible design, the target coordinate determination module 302 is specifically used to: use the first pending coordinates of the object to be measured in continuous video frames to determine the motion trajectory of the object to be measured during the movement, and determine the landing direction of the object to be measured in the image coordinate system corresponding to the predicted coordinate axis; confirm the landing point according to the coordinate value of the first pending coordinate on the predicted coordinate axis, and determine the first target coordinate.

[0046] In one possible design, the target coordinate determination module 302 is specifically used to: obtain the coordinate value under the predicted coordinate axis corresponding to each of the first pending coordinates, obtain the maximum coordinate value or the minimum coordinate value among the coordinate values ​​under the predicted coordinate axis; obtain the first pending coordinate corresponding to the maximum coordinate value or the minimum coordinate value, and use the first pending coordinate corresponding to the maximum coordinate value or the minimum coordinate value as the first target coordinate.

[0047] In a possible design, the target coordinate determination module 302 is specifically used to: determine the video frame to which the first target coordinates belong, and use the video frame to which the first target coordinates belong as the target video frame.

[0048] In one possible design, the coordinate transformation module 303 is specifically used to: obtain a coordinate transformation matrix between the preset image coordinate system and the plane coordinate system established based on overlooking the scoring area; and use the coordinate transformation matrix to transform the first range coordinates and the first target coordinates into corresponding second range coordinates and the second target coordinates.

[0049] In one possible design, the target area determination module 304 is specifically used to: determine, according to the second range coordinates, a second scoring area in a plane coordinate system established based on overlooking the scoring area; and determine the target area of ​​the landing point according to the position of the second target coordinates in the second scoring area.

[0050] The above device determines the first target coordinates of the object to be measured by the first undetermined coordinates of the object to be measured in the continuous video frames, so as to accurately obtain the target video frame of the landing point. In addition, according to the coordinate conversion rule, the first range coordinates and the first target coordinates on the target video frame are converted into the corresponding second range coordinates and the second target coordinates, so that the target area of ​​the landing point can be intuitively determined based on the second range coordinates and the second target coordinates based on the plane overlooking the landing area, thereby improving the accuracy of determining the target area of ​​the landing point.

[0051] See also Figure 4 , Figure 4 4 is a schematic diagram of a computer device provided in an embodiment of the present application, wherein the computer device 40 includes a processor 401 and a memory 402. The processor 401 is connected to the memory 402, for example, the processor 401 may be connected to the memory 402 via a bus.

[0052] The processor 401 is configured to support the computer device 40 to perform the corresponding functions in the method in the above method embodiment. The processor 401 can be a central processing unit (CPU), a network processor (NP), a hardware chip or any combination thereof. The above hardware chip can be an application specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof. The above PLD can be a complex programmable logic device (CPLD), a field programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof. Specifically, the processor 401 can be a processor in a single-chip microcomputer, and the memory 402 is used to store program code, etc. The memory 402 may include a volatile memory (VM), such as a random access memory (RAM); the memory 402 may also include a non-volatile memory (NVM), such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); the memory 402 may also include a combination of the above types of memory.

[0053] The processor 401 may call the program code to perform the following operations:

[0054] Acquire continuous video frames of the object to be measured, and determine first undetermined coordinates of the object to be measured according to the video frames, where the first undetermined coordinates are position coordinates of the object to be measured in each video frame in an image coordinate system;

[0055] Determine the first target coordinates of the object to be measured according to the first undetermined coordinates, and determine the corresponding target video frame according to the first target coordinates, wherein the first target coordinates are the position coordinates of the landing point of the object to be measured in the image coordinate system, and the target video frame is the video frame in which the landing point of the object to be measured is detected;

[0056] According to a coordinate conversion rule, the first range coordinates and the first target coordinates on the target video frame are converted into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on looking down at the scoring area, wherein the first range coordinates are coordinates of points determining the first scoring area based on the image coordinate system;

[0057] A target area of ​​the landing point is determined according to the second range coordinates and the second target coordinates, and the target area is a scoring area into which the object to be measured falls.

[0058] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a computer, the computer executes the method described in the above embodiment.

[0059] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0060] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A method for detecting a landing point area of ​​a space moving object, characterized in that: include: Acquire continuous video frames of the object to be measured, and determine first undetermined coordinates of the object to be measured according to the video frames, where the first undetermined coordinates are position coordinates of the object to be measured in each video frame in an image coordinate system; Determine the first target coordinates of the object to be measured according to the first undetermined coordinates, and determine the corresponding target video frame according to the first target coordinates, wherein the first target coordinates are the position coordinates of the landing point of the object to be measured in the image coordinate system, and the target video frame is the video frame in which the landing point of the object to be measured is detected; According to a coordinate conversion rule, the first range coordinates and the first target coordinates on the target video frame are converted into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on the overhead scoring area, wherein the first range coordinates are coordinates of points determining the first scoring area based on the image coordinate system; Determine a target area of ​​the landing point according to the second range coordinates and the second target coordinates, the target area being a scoring area into which the object to be measured falls; The determining the first target coordinates of the object to be measured according to the first undetermined coordinates includes: Determine the motion trajectory of the object to be measured during the movement process using the first undetermined coordinates of the object to be measured in the continuous video frames, and determine the predicted coordinate axis corresponding to the landing direction of the object to be measured in the image coordinate system; Confirm the landing point according to the coordinate value of the first undetermined coordinate on the predicted coordinate axis to determine the first target coordinate; The determining the first target coordinates according to the coordinate point on the predicted coordinate axis corresponding to the first undetermined coordinates includes: Obtaining a coordinate value under a predicted coordinate axis corresponding to each of the first undetermined coordinates, and obtaining a maximum coordinate value or a minimum coordinate value among the coordinate values ​​under the predicted coordinate axis; Acquire a first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value, and use the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value as the first target coordinate; The step of determining the first undetermined coordinates of the object to be measured according to the video frame includes: The first undetermined coordinates of the object to be detected in the video frame are detected by a convolutional neural network target detection algorithm; the convolutional neural network target detection algorithm consists of a backbone network, a path aggregation network and an output segment.

2. A method for detecting a landing point area of ​​a space moving object according to claim 1, characterized in that: The step of converting the first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established by looking down at the scoring area according to a coordinate conversion rule includes: Acquire a coordinate conversion matrix between the preset image coordinate system and a plane coordinate system established based on looking down at the scoring area; The first range coordinates and the first target coordinates are converted into corresponding second range coordinates and the second target coordinates using the coordinate conversion matrix.

3. The method for detecting the landing point area of ​​a space moving object according to claim 1, characterized in that: The determining the target area of ​​the landing point according to the second range coordinates and the second target coordinates includes: Determine, according to the second range coordinates, a second scoring area in a plane coordinate system established based on the overhead view of the scoring area; A target area for the landing point is determined according to the position of the second target coordinates in the second scoring area.

4. The method for detecting the landing point area of ​​a space moving object according to claim 1, characterized in that: The determining the corresponding target video frame according to the first target coordinates includes: The video frame to which the first target coordinates belong is determined, and the video frame to which the first target coordinates belong is used as the target video frame.

5. A device for detecting a landing point area of ​​a space moving object, characterized in that: include: A module for acquiring coordinates to be determined, used for acquiring continuous video frames of the object to be measured, and determining first coordinates to be determined of the object to be measured according to the video frames; Detecting the first undetermined coordinates of the object to be detected in the video frame by a convolutional neural network target detection algorithm; the convolutional neural network target detection algorithm is composed of a backbone network, a path aggregation network and an output segment; a target coordinate determination module, used to determine the first target coordinates of the object to be measured according to the first undetermined coordinates, and determine the corresponding target video frame according to the first target coordinates, wherein the first target coordinates are the position coordinates of the landing point of the object to be measured in the image coordinate system, and the target video frame is the video frame in which the landing point of the object to be measured is detected; Determine the motion trajectory of the object to be measured during the movement by using the first undetermined coordinates of the object to be measured in the continuous video frames, and determine the predicted coordinate axis corresponding to the landing direction of the object to be measured in the image coordinate system; confirm the landing point according to the coordinate value of the first undetermined coordinate on the predicted coordinate axis, and determine the first target coordinates; obtain the coordinate value under the predicted coordinate axis corresponding to each of the first undetermined coordinates, and obtain the maximum coordinate value or the minimum coordinate value among the coordinate values ​​under the predicted coordinate axis; obtain the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value, and use the first undetermined coordinate corresponding to the maximum coordinate value or the minimum coordinate value as the first target coordinate; A coordinate conversion module, configured to convert the first range coordinates and the first target coordinates on the target video frame into corresponding second range coordinates and second target coordinates in a plane coordinate system established based on a bird's-eye view of the scoring area according to a coordinate conversion rule, wherein the first range coordinates are coordinates of points determining the first scoring area based on an image coordinate system; A target area determination module is used to determine the target area of ​​the landing point according to the second range coordinates and the second target coordinates, and the target area is the scoring area where the object to be measured falls.

6. A computer device, characterized in that: The method comprises a memory and one or more processors, wherein the one or more processors are used to execute one or more computer programs stored in the memory, and when the one or more processors execute the one or more computer programs, the computer device implements the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 4.