A method and storage medium for long-range identification of moving targets based on a flight platform

Abstract

A method and storage medium for long-range recognition of moving targets based on a flight platform includes: detecting targets in an image using a target detection network; performing pipeline association on the targets detected in each frame and updating the positions of the targets in the pipeline; determining keyframe images and mapping each subsequent frame to the keyframe image, calculating the perspective transformation matrix of each subsequent frame relative to the keyframe image; mapping the center point of each detection box in each subsequent frame to the keyframe image based on the calculated perspective transformation matrix; comparing the target mapping result of the current frame image in the same pipeline with the target mapping result of the previous frame image in the same pipeline based on the pipeline association result, confirming the target motion state; setting the pipeline of the moving target to a moving state, setting the pipeline of the stationary target to a stationary state, and outputting the final determined position of the target in the moving pipeline, thereby recognizing and tracking the moving target.

Description

Technical Field

[0001] This invention relates to the field of image recognition and tracking, specifically to a method and storage medium for long-range recognition of moving targets based on a flight platform. Background Technology

[0002] In missions such as military reconnaissance, search and rescue, or environmental monitoring, rapid and accurate target location is often crucial for mission success. For moving targets, failure to identify and track them in a timely manner can cause them to move out of the field of view, leading to mission failure. Furthermore, long-range target identification and tracking also helps improve the real-time nature and accuracy of information acquisition.

[0003] Existing target detection technologies based on target detection networks utilize information from single-frame images to detect pre-trained targets in a scene. However, they cannot determine the motion state of the target or distinguish between moving and stationary targets. Furthermore, these methods only utilize information from single-frame images of the target, resulting in poor reliability. The method of identifying targets by accumulating confidence scores leads to slow recognition speed and short range.

[0004] Existing moving target detection methods based on optical flow, frame difference, and background modeling are computationally intensive and time-consuming, which is not suitable for scenarios with high real-time requirements. In addition, while these moving target detection methods may perform well on stationary platforms, they often detect stationary backgrounds as moving targets on flying platforms due to the platform's own motion, resulting in a high false alarm rate. Summary of the Invention

[0005] In view of the technical defects and drawbacks existing in the prior art, embodiments of the present invention provide a method and storage medium for long-range identification of moving targets based on a flight platform to overcome the above problems or at least partially solve the above problems. The specific solution is as follows:

[0006] As a first aspect of the present invention, a method for long-range identification of moving targets based on a flight platform is provided, the method comprising:

[0007] When the flight platform acquires images containing both targets and scenes, a target detection network is used to detect targets in the images.

[0008] For each frame of image, the detected targets are associated with pipelines, and the positions of the targets in the pipelines are updated;

[0009] Determine the keyframe image, map each subsequent frame image to the keyframe image, and calculate the perspective transformation matrix of each subsequent frame image relative to the keyframe image.

[0010] Based on the calculated perspective transformation matrix, the center point of each detection box in each subsequent frame image is mapped to the keyframe image;

[0011] Based on the pipeline association results, the target mapping results of the current frame image in the same pipeline are compared with the target mapping results of the previous frame image to confirm the target motion state;

[0012] Set the pipes of moving targets to a moving state and the pipes of stationary targets to a stationary state, and output the final determined position of the pipe target in a moving state to identify and track the moving targets.

[0013] Furthermore, when acquiring images containing targets and scenes on the flight platform, the detection of targets in the images using a target detection network includes:

[0014] Step 101: Collect images containing the target and the scene, label the target, and save the images to the established target image database;

[0015] Step 102: Input the images from the target image database into the target detection network for training. After the target detection network is trained, convert the target detection network into a model format supported by the ARM hardware platform.

[0016] Step 103: Load the target detection model onto the ARM hardware platform, perform inference on the ARM hardware platform, and detect all targets in the current frame image.

[0017] Furthermore, the step of performing pipeline association on the targets detected in each frame of the image and updating the positions of the targets in the pipeline includes:

[0018] Step 201: For the first frame of the detected image, initialize a pipeline for each detection result, and initialize a tracking chain using the target detection position in each pipeline;

[0019] Step 202: In each subsequent frame, first determine whether the number of target pipes in the previous frame is greater than 0. If the number of target pipes is greater than 0, then use the tracking chain in each pipe to predict the position of the target in the corresponding pipe in the current frame.

[0020] Step 203: Calculate the IoU value of the predicted bounding box of the target in the current frame image and all detection results in each pipeline;

[0021] Step 204: Based on the IOU value, use Hungarian matching to obtain the matching relationship between each pipeline target prediction box and all detection boxes, and update the position of the target in the pipeline and the tracking chain.

[0022] Further, step 204 includes:

[0023] a. If the detection box does not match any of the target channels, initialize a channel for the detection box and initialize a tracking chain based on the detection results;

[0024] b. If the detection box successfully matches a target pipeline, the position of the target in the pipeline and the tracking chain are updated using the detection result;

[0025] c. If the target pipeline does not match any detection boxes, the position of the target in the pipeline is updated using the prediction results, and the pipeline is marked as not matching the current frame image.

[0026] Furthermore, in step 202, if the number of pipes is not greater than 0, the operation in step 201 is repeated to initialize one pipe for each detection result.

[0027] Further, the step of determining the keyframe image, mapping each subsequent frame image to the keyframe image, and calculating the perspective transformation matrix of each subsequent frame image relative to the keyframe image specifically includes:

[0028] Step 301: In the first frame of the detected image, the image is divided into four quadrants with the center point as the boundary. A preset size image block is randomly selected from each of the four quadrants of the image for tracking initialization. This frame image is the key frame image.

[0029] Step 302: In each subsequent frame, track the four image blocks selected in the keyframe image, and calculate the perspective transformation matrix between the current frame image and the keyframe image using the center points of the four tracking boxes in the current frame image and the center points of the four tracking boxes in the keyframe image.

[0030] Furthermore, the method also includes: determining whether the keyframe image needs to be updated. There are three conditions for updating the keyframe image; if any one of the following three conditions is met, the keyframe image needs to be updated:

[0031] Condition 1: The tracking confidence of the four image blocks in the current frame is lower than the confidence threshold;

[0032] Condition 2: The positions of the four image blocks tracked in the current frame are located at the edges of the image;

[0033] Condition 3: The maximum IoU value between the tracking boxes of the four image blocks in the current frame and all detection boxes in the current frame is greater than 0.

[0034] Further, mapping the center point of each detection box in each subsequent frame image to the keyframe image based on the calculated perspective transformation matrix includes:

[0035] If the frame image is a keyframe image, no mapping is needed; the coordinates of the target mapping point in the pipeline are updated using the coordinates of the center point of the detection box.

[0036] If the current frame image is not a keyframe image, the center point of the detection box is calculated, and the coordinates of the center point mapped to the keyframe image are calculated using the perspective transformation matrix. The coordinates of the target mapping point in the pipeline are then updated.

[0037] Furthermore, the step of comparing the target mapping result of the current frame image in the same pipeline with the target mapping result of the previous frame image in the same pipeline based on the pipeline association result to confirm the target motion state includes:

[0038] Step 501: Assume the coordinates of a target mapping point in the current frame image are (x1, y1), and the coordinates of the corresponding target mapping point in the previous frame image are (x2, y2). If x1 - x2 > 0, the target in the current frame image moves in the positive x direction, the x-direction movement count is incremented by 1, the x-direction movement count is 0, and the x-direction movement displacement is (x1 - x2). If y1 - y2 > 0, the y-direction movement count is incremented by 1, the y-direction movement count is 0, and the y-direction movement displacement is (y1 - y2).

[0039] Step 502: Within 1 second, the cumulative count of the target's forward and reverse x-direction movements is calculated, and the cumulative displacement of the x-direction movements is added together. The cumulative count of the target's forward and reverse y-direction movements is calculated, and the cumulative displacement of the y-direction movements is added together.

[0040] Step 503: If the cumulative count of movement in the positive x direction is greater than the calculation threshold for the positive x direction and the cumulative count of movement in the negative x direction is 0, and the displacement in the x direction is greater than the displacement threshold in the x direction, then the target moves in the positive x direction and the x direction movement flag is set to 1.

[0041] If the cumulative count of movement in the opposite x direction is greater than the x-direction count threshold and the cumulative count of movement in the positive x direction is 0, and the x-direction displacement is greater than the x-direction displacement threshold, then the target moves in the opposite x direction and the x-direction movement flag is set to 1.

[0042] If the cumulative count of movement in the positive y direction is greater than the y-direction calculation threshold and the cumulative count of movement in the negative y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the positive y direction and the y-direction movement flag is set to 1.

[0043] If the cumulative count of movement in the opposite y direction is greater than the y-reverse calculation threshold and the cumulative count of movement in the positive y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the opposite y direction, and the y-direction movement flag is set to 1.

[0044] Step 504: If the x-direction motion flag or the y-direction motion flag is set to 1 in step 503, then the corresponding target is considered to be a moving target; otherwise, the corresponding target is considered to be a stationary target.

[0045] As a second aspect of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a computer, the computer program causes the computer to perform the long-range moving target identification method based on a flight platform as described above.

[0046] The present invention has the following beneficial effects:

[0047] (1) In the task of identifying and tracking moving targets, for multi-target scenarios, it can quickly and accurately identify moving targets and accurately track them.

[0048] (2) It can achieve ultra-long-distance recognition of moving targets and prevent targets from moving out of the field of view.

[0049] (3) When registering non-keyframe images with keyframe images, the present invention adopts the method of tracking four image blocks in the background for registration, which avoids the complicated process of conventional feature point matching and effectively reduces the time consumption of moving target detection.

[0050] (4) This technology utilizes the temporal and spatial continuity of the target in the image sequence to effectively improve the recall and accuracy of moving target detection. Attached Figure Description

[0051] Figure 1 A flowchart of a method for long-range identification of moving targets based on a flight platform provided in an embodiment of the present invention;

[0052] Figure 2 A flowchart of target detection provided for embodiments of the present invention;

[0053] Figure 3 A flowchart illustrating the association of target pipelines provided in this embodiment of the invention;

[0054] Figure 4 This is a flowchart of the image registration process provided in an embodiment of the present invention;

[0055] Figure 5 A flowchart for target motion attribute recognition provided for embodiments of the invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] See Figure 1This invention provides a method for long-range identification of moving targets based on a flight platform, the method comprising:

[0058] Step 1: When the flight platform acquires images containing both the target and the scene, a target detection network is used to detect the targets in the images.

[0059] Step 2: Perform pipeline association on the targets detected in each frame of the image and update the position of the targets in the pipeline;

[0060] Step 3: Determine the keyframe image and map each subsequent frame image to the keyframe image, and calculate the perspective transformation matrix of each subsequent frame image relative to the keyframe image.

[0061] This step is used to address the impact of platform motion. Optionally, this aspect selects four regions in the keyframe image for tracking initialization, and each subsequent frame image tracks these four regions separately. The perspective transformation matrix of each subsequent frame image relative to the keyframe image is calculated using the center point positions of the four tracked regions.

[0062] Step 4: Based on the calculated perspective transformation matrix, map the center point of each detection box in each subsequent frame image to the keyframe image. The coordinates of the center point of the detection box in the keyframe image are the coordinates of the target mapping point.

[0063] Step 5: Based on the pipeline association results, compare the target mapping results of the current frame image in the same pipeline with the target mapping results of the previous frame or multiple frames of the current frame image to confirm the target motion state.

[0064] For example, the result of the target mapping of the current frame image in the same pipeline is compared with the result of the target mapping of the corresponding target in the previous frame image. The motion direction and displacement of the target in the current frame image relative to the target in the previous frame image are counted. After the accumulation of multiple frames, if the target moves a certain displacement in a certain direction in multiple consecutive frames, the target is considered to be a moving target; otherwise, the target is considered to be a stationary target.

[0065] Step 6: Set the pipe of the moving target to a moving state, set the pipe of the stationary target to a stationary state, and output the final determined position of the pipe target in the moving state to identify and track the moving target.

[0066] This invention uses image-detected targets for pipeline association. In the task of identifying and tracking moving targets, it can quickly and accurately identify moving targets and perform accurate tracking in multi-target scenarios. When registering non-keyframe images with keyframe images, this invention uses a method of tracking four image blocks in the background for registration, avoiding the complex process of conventional feature point matching and effectively reducing the time consumption of moving target detection.

[0067] See Figure 2 As a specific embodiment of the present invention, the step of performing pipeline association on the targets detected in each frame of the image and updating the position of the targets in the pipeline specifically includes:

[0068] Step 101: Collect images containing the target and the scene, label the target, and save the images to the established target image database;

[0069] Step 102: Input the images from the target image database into the target detection network for training. After the target detection network is trained, convert the target detection network into a model format supported by the ARM hardware platform.

[0070] Step 103: Load the target detection model onto the ARM hardware platform, perform inference on the ARM hardware platform, and detect all targets in the current frame image.

[0071] See Figure 3 As a specific embodiment of the present invention, the step of performing pipeline association on the targets detected in each frame of the image and updating the position of the targets in the pipeline specifically includes:

[0072] Step 201: For the first frame of the detected image, initialize a pipeline for each detected target, and initialize a tracking chain using the position of the detected target in each pipeline.

[0073] Step 202: In each subsequent frame, first determine whether the number of target pipes in the previous frame is greater than 0. If the number of target pipes is greater than 0, then use the tracking chain in each pipe to predict the position of the target in the current frame. If the number of pipes is not greater than 0, then repeat the operation in step 201 to initialize a pipe for each detected target.

[0074] Step 203: Calculate the IoU value between the predicted bounding box of the target in the current frame image and all detection boxes in each pipeline.

[0075] Step 204: Based on the IOU value, use Hungarian matching to obtain the matching relationship between each pipeline target prediction box and all detection boxes. Three matching relationships will appear, as follows:

[0076] a) If the detection box does not match any of the target channels, initialize a channel for the detection box and initialize a tracking chain based on the detection results;

[0077] b) If the detection box successfully matches a target pipeline, the position of the target in the pipeline and the tracking chain are updated using the detection result;

[0078] c) If the target pipeline does not match any detection boxes, the position of the target in the pipeline is updated using the prediction results, and the pipeline is marked as not matching the current frame image.

[0079] Step 205, pipeline connection complete.

[0080] See Figure 4 As a specific embodiment of the present invention, the step of determining the keyframe image, mapping each subsequent frame image to the keyframe image, and calculating the perspective transformation matrix of each subsequent frame image relative to the keyframe image specifically includes:

[0081] Step 301: In the first frame of the detected image, the image is divided into four quadrants with the center point of the image as the boundary. A 20*15 image block is randomly selected from each of the four quadrants of the image for tracking initialization. This frame image is the key frame image. Each image block is tracked by a tracking box.

[0082] Step 302: In each subsequent frame, the four image blocks selected in the keyframe image are first tracked. The perspective transformation matrix between the current frame image and the keyframe image is calculated using the center points of the four tracking boxes in the current frame image and the center points of the four tracking boxes in the keyframe image.

[0083] Optionally, the method further includes: determining whether the next frame image needs to be updated with a keyframe image. There are three conditions for updating the keyframe image, and if any one of the following three conditions is met, the keyframe image needs to be updated:

[0084] Condition 1: The tracking confidence of the four image blocks in the current frame is lower than the confidence threshold;

[0085] Condition 2: The positions of the four image blocks tracked in the current frame are located at the edges of the image;

[0086] Condition 3: The maximum IoU value between the tracking boxes of the four image blocks in the current frame and all detection boxes in the current frame is greater than 0.

[0087] If any one of the above three conditions is met, the keyframe image needs to be updated. The current frame image or the previous frame image is updated to the keyframe image according to the method in step 301.

[0088] As a specific embodiment of the present invention, the step of performing pipeline association on the targets detected in each frame of the image and updating the position of the targets in the pipeline specifically includes:

[0089] Step 401: If the frame image is a keyframe image, no mapping is required. The coordinates of the target mapping point in the pipeline are updated using the center point coordinates of the detection box.

[0090] Step 402: If the current frame image is not a keyframe image, calculate the center point of the detection box, and use the perspective transformation matrix to calculate the coordinates of the center point mapped to the keyframe image, and update the coordinates of the target mapping point in the pipeline.

[0091] See Figure 5 As a specific embodiment of the present invention, the step of comparing the target mapping result of the current frame image in the same pipeline with the target mapping result of the previous frame image in the same pipeline based on the pipeline association result to confirm the target motion state includes:

[0092] Step 501: Assume the coordinates of a target mapping point in the current frame image are (x1, y1), and the coordinates of the corresponding target mapping point in the previous frame image are (x2, y2). If x1 - x2 > 0, then the target moves in the positive x direction in the current frame image, the x-direction movement count is incremented by 1, the x-direction movement count is 0, and the x-direction movement displacement is (x1 - x2). If y1 - y2 > 0, then the y-direction movement count is incremented by 1, the y-direction movement count is 0, and the y-direction movement displacement is (y1 - y2).

[0093] Step 502: Within 1 second, the cumulative count of the target's forward and reverse x-direction movements is calculated, and the cumulative displacement of the x-direction movements is added together. The cumulative count of the target's forward and reverse y-direction movements is calculated, and the cumulative displacement of the y-direction movements is added together.

[0094] Step 503: If the cumulative count of movement in the positive x direction is greater than the calculation threshold for the positive x direction and the cumulative count of movement in the negative x direction is 0, and the displacement in the x direction is greater than the displacement threshold in the x direction, then the target moves in the positive x direction and the x direction movement flag is set to 1.

[0095] If the cumulative count of movement in the opposite x direction is greater than the x-direction count threshold and the cumulative count of movement in the positive x direction is 0, and the x-direction displacement is greater than the x-direction displacement threshold, then the target moves in the opposite x direction and the x-direction movement flag is set to 1.

[0096] If the cumulative count of movement in the positive y direction is greater than the y-direction calculation threshold and the cumulative count of movement in the negative y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the positive y direction and the y-direction movement flag is set to 1.

[0097] If the cumulative count of movement in the opposite y direction is greater than the y-reverse calculation threshold and the cumulative count of movement in the positive y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the opposite y direction and the y-direction movement flag is set to 1.

[0098] Step 504: If the x-direction motion flag or the y-direction motion flag in step 503 is set to 1, then the target is considered a moving target and the pipe where the target is located is set to a moving pipe; otherwise, the target is considered a stationary target and the pipe where the target is located is set to a stationary pipe.

[0099] As another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and when the computer program is executed by a computer, the computer performs the long-range moving target identification method based on the flight platform as described above.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for long-range identification of moving targets based on a flight platform, characterized in that, The method includes: When the flight platform acquires images containing both targets and scenes, a target detection network is used to detect targets in the images. For each frame of image, the detected targets are associated with pipelines, and the positions of the targets in the pipelines are updated; Determine the keyframe image, map each subsequent frame image to the keyframe image, and calculate the perspective transformation matrix of each subsequent frame image relative to the keyframe image. Based on the calculated perspective transformation matrix, the center point of each detection box in each subsequent frame image is mapped to the keyframe image; Based on the pipeline association results, the target mapping result of the current frame image in the same pipeline is compared with the target mapping result of the previous frame image to confirm the target motion state; Set the pipe of the moving target to a moving state, set the pipe of the stationary target to a stationary state, and output the final determined position of the pipe target in the moving state to identify and track the moving target. The step of performing pipeline association on the targets detected in each frame of the image and updating the positions of the targets in the pipeline includes: Step 201: For the first frame of the detected image, initialize a pipeline for each detection result, and initialize a tracking chain using the target detection position in each pipeline; Step 202: In each subsequent frame, first determine whether the number of target pipes in the previous frame is greater than 0. If the number of target pipes is greater than 0, then use the tracking chain in each pipe to predict the position of the target in the corresponding pipe in the current frame. Step 203: Calculate the IoU value between the predicted bounding box of the target in the current frame image and all detected bounding boxes in each pipeline; Step 204: Based on the IOU value, use Hungarian matching to obtain the matching relationship between each pipeline target prediction box and all detection boxes, and update the position of the target in the pipeline and the tracking chain. Specifically, determining the keyframe image, mapping each subsequent frame image to the keyframe image, and calculating the perspective transformation matrix of each subsequent frame image relative to the keyframe image include: Step 301: In the first frame of the detected image, the image is divided into four quadrants with the center point of the image as the boundary. A preset size image block is randomly selected in each of the four quadrants of the image for tracking initialization, and this frame of the image is used as the key frame image. Step 302: In each subsequent frame, track the four image blocks selected in the keyframe image, and calculate the perspective transformation matrix between the current frame image and the keyframe image using the center points of the four tracking boxes in the current frame image and the center points of the four tracking boxes in the keyframe image.

2. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, When acquiring images containing targets and scenes on the flight platform, the detection of targets in the images using a target detection network includes: Step 101: Collect images containing the target and the scene, label the target, and save the images to the established target image database; Step 102: Input the images from the target image database into the target detection network for training. After the target detection network is trained, convert the target detection network into a model format supported by the ARM hardware platform. Step 103: Load the target detection model onto the ARM hardware platform, perform inference on the ARM hardware platform, and detect all targets in the current frame image.

3. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, Step 204 includes: a. If the detection box does not match any of the target channels, initialize a channel for the detection box and initialize a tracking chain based on the detection results; b. If the detection box successfully matches a target pipeline, the position of the target in the pipeline and the tracking chain are updated using the detection result; c. If the target pipeline does not match any detection boxes, the position of the target in the pipeline is updated using the prediction results, and the pipeline is marked as not matching the current frame image.

4. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, In step 202, if the number of pipes is not greater than 0, repeat the operation in step 201 to initialize one pipe for each detection result.

5. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, The method further includes: determining whether the keyframe image needs to be updated. There are three conditions for updating the keyframe image. If any one of the following three conditions is met, the keyframe image needs to be updated: Condition 1: The tracking confidence of the four image blocks in the current frame is lower than the confidence threshold; Condition 2: The positions of the four image blocks tracked in the current frame are located at the edges of the image; Condition 3: The maximum IoU value between the tracking boxes of the four image blocks in the current frame and all detection boxes in the current frame is greater than 0.

6. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, The step of mapping the center point of each detection box in each subsequent frame image to the keyframe image based on the calculated perspective transformation matrix includes: If the current frame image is a keyframe image, no mapping is required; the coordinates of the target mapping point in the pipeline are updated using the center point coordinates of the detection box. If the current frame image is not a keyframe image, the center point of the detection box is calculated, and the coordinates of the center point mapped to the keyframe image are calculated using the perspective transformation matrix. The coordinates of the target mapping point in the pipeline are then updated.

7. The method for long-range identification of moving targets based on a flight platform according to claim 1, characterized in that, The step of comparing the target mapping result of the current frame image in the same pipeline with the target mapping result of the previous frame image in the same pipeline based on the pipeline association result to confirm the target motion state includes: Step 501: Assume the coordinates of a pipe target mapping point in the current frame image are (x1, y1), and the coordinates of the corresponding pipe target mapping point in the previous frame image are (x2, y2). If x1 - x2 > 0, then the target in the current frame image moves in the positive x direction, incrementing the x-direction motion count by 1, and decrementing the x-direction motion count by 0, with a displacement of (x1 - x2). If y1 - y2 > 0, then the y-direction motion count increments by 1, decrementing the y-direction motion count by 0, with a displacement of (y1 - y2). Step 502: Within 1 second, the cumulative count of the target's forward and reverse x-direction movements is calculated, and the cumulative displacement of the x-direction movements is added together. The cumulative count of the target's forward and reverse y-direction movements is calculated, and the cumulative displacement of the y-direction movements is added together. Step 503: If the cumulative count of movement in the positive x direction is greater than the calculation threshold for the positive x direction and the cumulative count of movement in the negative x direction is 0, and the displacement in the x direction is greater than the displacement threshold in the x direction, then the target moves in the positive x direction and the x direction movement flag is set to 1. If the cumulative count of movement in the opposite x direction is greater than the x-direction count threshold and the cumulative count of movement in the positive x direction is 0, and the x-direction displacement is greater than the x-direction displacement threshold, then the target moves in the opposite x direction and the x-direction movement flag is set to 1. If the cumulative count of movement in the positive y direction is greater than the y-direction calculation threshold and the cumulative count of movement in the negative y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the positive y direction and the y-direction movement flag is set to 1. If the cumulative count of movement in the opposite y direction is greater than the y-reverse calculation threshold and the cumulative count of movement in the positive y direction is 0, and the displacement in the y direction is greater than the y-direction displacement threshold, then the target moves in the opposite y direction and the y-direction movement flag is set to 1. Step 504: If the x-direction motion flag or the y-direction motion flag is set to 1 in step 503, then the corresponding target is considered to be a moving target; otherwise, the corresponding target is considered to be a stationary target.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a computer, causes the computer to perform the long-range moving target identification method based on a flight platform as described in any one of claims 1 to 7.

Images