Anti-interference rotating target tracking method and embedded device

Abstract

The invention relates to a target tracking method and embedded equipment, in particular to an anti-interference rotating target tracking method and embedded equipment, and solves the problems that an existing method cannot accurately track a rotating target and is complex in calculation. The method comprises the following steps: step 1, acquiring a first frame image block; 2, affine rotation is carried out; step 3, acquiring a rotation angle alpha of a second frame target; 4, obtaining the final position of the second frame target; step 5, acquiring a second frame as a current frame; step 6, acquiring a current frame image block; 7, affine rotation is carried out; 8, acquiring a rotation angle alpha of the next frame of target; step 9, judging; if not, executing the step 10, and if yes, executing the step 11; 10, the response diagram position serves as the final position updating parameter of the next frame of target, and the step 12 is executed; step 11, fusing Kalman filtering to obtain the final position of the next frame of target; step 12, judging; and if not, returning to the step 6.

Description

Technical Field

[0001] The invention relates to a target tracking method and an embedded device, and in particular to an interference-resistant rotating target tracking method and an embedded device. Background Art

[0002] Tracking ground targets through remote sensing satellites has important application value in disaster detection, traffic monitoring, and marine protection.

[0003] When tracking ground targets through remote sensing satellites, there are many existing tracking methods that can be used, but most of them cannot accurately track rotating targets. Although the Siamese R-CNN tracking method (a neural network tracking method for visual tracking through re-detection) can achieve relatively accurate tracking of rotating targets, it requires a large amount of data training during the tracking process, and the calculation process is very complicated, and it cannot be installed and run on embedded devices. However, due to the bird's-eye view of remote sensing observation, it is very common for the target to be tracked to be occluded and rotated. Therefore, it is urgent to develop a rotating target tracking method that has a simple calculation process, can be installed and run on embedded devices, and can accurately track occluded and rotating targets. Summary of the invention

[0004] The purpose of the present invention is to solve the technical problem that most of the existing tracking methods cannot achieve accurate tracking when tracking occluded and rotating targets, and some can achieve relatively accurate tracking but the calculation process is too complicated and cannot be installed and run on embedded devices, and to provide an interference-resistant rotating target tracking method and embedded device.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for tracking a rotating target with anti-interference is characterized in that it comprises the following steps: Step 1: Obtain a first frame image, frame the position of the target to be tracked on the image, and obtain an image block of the target to be tracked on the first frame image; Step 2: Performing affine rotations of n pairs of positive and negative symmetric angles on the image blocks obtained in step 1 to obtain 2n rotated image blocks, where n is a natural number greater than 1; Step 3: According to the image block obtained in step 1 and the 2n rotated image blocks obtained in step 2, a maximum target response map of the second frame image is obtained by using a kernel correlation filter model; and then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2, a predicted rotation angle α of the target to be tracked in the second frame image is obtained; the rotation angle α of the target to be tracked refers to the rotation angle of the target to be tracked relative to the target to be tracked in the previous frame image; Step 4: taking the position corresponding to the maximum target response map of the second frame image obtained in step 3 as the predicted final position of the target to be tracked in the second frame image, and updating the parameters of the kernel correlation filter model in step 3 based on the final position; Step 5: Obtain the second frame image as the current frame image; Step 6: based on the final position of the target to be tracked in the current frame image predicted by the previous frame image, an image block of the target to be tracked on the current frame image is obtained; Step 7: For the image block obtained in step 6, perform affine rotation of the angle α of the target to be tracked in the current frame image predicted based on the previous frame image, and the angle β in the nearest pool corresponding to the angle α, respectively, to obtain two rotated image blocks; the angle β in the nearest pool is the angle value closest to the angle α and having an absolute value greater than the absolute value of the angle α among the n pairs of positive and negative symmetric angles set in step 2, except the angle α. When the angle α is equal to the angle with the largest absolute value among the n pairs of positive and negative symmetric angles set in step 2, the angle β in the nearest pool is the angle value closest to the angle α except the angle α. Step 8: According to the image block obtained in step 6 and the two rotated image blocks obtained in step 7, the maximum target response map of the next frame image is obtained by using the kernel correlation filter model whose parameters have been updated most recently; and then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles when the two rotated image blocks are obtained in step 7, the predicted rotation angle α of the target to be tracked in the next frame image is obtained; Step 9: According to the preset occlusion judgment standard of the target to be tracked, determine whether the target to be tracked is occluded; if not, execute step 10; if yes, execute step 11; Step 10: taking the position corresponding to the maximum target response map of the next frame image obtained in step 8 as the predicted final position of the target to be tracked in the next frame image, and based on the final position, updating the parameters of the kernel correlation filter model most recently used in step 8; then executing step 12; Step 11: Based on the maximum target response map of the next frame image obtained in step 8 and the motion displacement of the target to be tracked in the most recent M frames, the Kalman filter is integrated to obtain the predicted final position of the target to be tracked in the next frame image, where M is an integer greater than or equal to 3; if the total number of frames before the current frame image is less than M frames, the motion displacement of the target to be tracked from the first frame image to the current frame image is obtained by integrating the Kalman filter; Step 12: Determine whether an end tracking instruction is received; if not, obtain the next frame image as the new current frame image, update the current frame image, and return to step 6; if yes, the tracking is completed.

[0006] Furthermore, in step 1, the position of the target to be tracked is framed by manually framing the position of the target to be tracked or by introducing a lightweight deep learning neural network to detect the target to be tracked.

[0007] Furthermore, in step 1 and step 6, the area of ​​the image block of the target to be tracked is 2.5 to 3.5 times the area of ​​the target to be tracked, and its specific value is determined by the size of the target to be tracked and is an empirical value; In step 2, the absolute values ​​of the set n pairs of positive and negative symmetry angles are all greater than zero and less than or equal to 3 degrees.

[0008] Furthermore, in order to comprehensively cover the rotation angles of the target to be tracked between frames and to simplify the calculation process and achieve obvious image affine rotation effects, in step 2, n is less than or equal to 4; the set n pairs of positive and negative symmetric angles are selected from ±0.5 degrees, ±1 degrees, ±1.5 degrees and ±2 degrees.

[0009] Further, in step 3, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the second frame image is: directly using the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 as the obtained predicted rotation angle α of the target to be tracked in the second frame image; In step 8, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the next frame image is: among the affine rotation angles of the two rotated image blocks obtained in step 7, the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image is directly used as the obtained predicted rotation angle α of the target to be tracked in the next frame image.

[0010] Alternatively, in order to make the predicted rotation angle α of the target to be tracked more accurate, and thus to achieve more accurate tracking of the rotating target, in step 3, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the second frame image is: The affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 is used as the theoretical value of the rotation angle α of the target to be tracked in the predicted second frame image, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as the measurement value of the rotation angle α of the target to be tracked in the predicted second frame image, and then the theoretical value and the measured value are averaged to obtain the rotation angle α of the target to be tracked in the predicted second frame image; In step 8, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the next frame image is: The affine rotation angle of the rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles of the two rotated image blocks obtained in step 7 is used as the theoretical value of the rotation angle α of the target to be tracked in the predicted next frame image, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as the measured value of the rotation angle α of the target to be tracked in the predicted next frame image, and then the theoretical value and the measured value are averaged to obtain the rotation angle α of the target to be tracked in the predicted next frame image.

[0011] Furthermore, in step 9, according to the preset occlusion judgment standard of the target to be tracked, the specific method for judging whether the target to be tracked is occluded is: The judgment is made according to the magnitude relationship between the peak value of the maximum target response map of the next frame image obtained in step 8 and the set occlusion detection threshold: when the peak value of the maximum target response map of the next frame image obtained in step 8 is greater than the set occlusion detection threshold, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded; The set occlusion detection threshold is 0.3-0.5, and its specific value is determined by the size and motion characteristics of the target to be tracked, and is an empirical value.

[0012] Alternatively, in step 9, according to a preset occlusion judgment standard of the target to be tracked, a specific method for judging whether the target to be tracked is occluded is: A lightweight deep learning neural network is introduced to detect the target to be tracked, and a judgment is made based on the detection results: if the target to be tracked can be detected and the confidence level is greater than 0.5, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded.

[0013] Furthermore, in order to make the predicted final position of the target to be tracked in the next frame image more accurate when the target to be tracked is occluded, thereby achieving more accurate tracking, and to make the calculation process relatively simple, in step 11, M is 3 to 5, and its specific value is determined by the motion characteristics of the target to be tracked and is an empirical value.

[0014] At the same time, the present invention also provides an embedded device, including a memory, a processor and a computer program stored in the memory; the special feature is that the processor executes the computer program to implement the steps of the above-mentioned anti-interference rotating target tracking method.

[0015] The beneficial effects of the present invention are: (1) The anti-interference rotating target tracking method of the present invention performs affine rotations of multiple pairs of positive and negative symmetric angles on the image blocks of the target to be tracked on the first frame image to obtain multiple rotated image blocks, and uses a kernel correlation filter model to obtain the maximum target response map of the second frame image based on the image blocks before rotation and the multiple rotated image blocks, thereby obtaining the predicted rotation angle α of the target to be tracked in the second frame image and the predicted final position of the target to be tracked in the second frame image, and updating the parameters of the kernel correlation filter model; then, the image blocks of the target to be tracked on the subsequent frame images are all subjected to the rotation angle α of the target to be tracked in the current frame image predicted based on the previous frame image, and the maximum target response map corresponding to the angle α. Affine rotation of the angle β in the near pool is performed to obtain two rotated image blocks. According to the image block before rotation and the two rotated image blocks, the kernel correlation filter model whose parameters have been updated most recently is used to obtain the maximum target response map of the next frame image, and then after obtaining the rotation angle α of the target to be tracked in the predicted next frame image, it is first judged whether the target to be tracked is occluded. Then, according to the judgment result, different methods are used to obtain the final position of the target to be tracked in the predicted next frame image, and when it is judged that no occlusion occurs, the parameters of the kernel correlation filter model are updated, and when it is judged that occlusion occurs, the parameters of the kernel correlation filter model are not updated until it is judged that no occlusion occurs. From the above description of the acquisition process, it can be seen that the present invention can not only obtain the final position of the target to be tracked in the predicted next frame image, but also obtain the rotation angle α of the target to be tracked in the predicted next frame image, so it can realize accurate tracking of the rotating target and improve the robustness of the kernel correlation filter model; in addition, for subsequent frame images other than the first frame, in the process of obtaining the final position of the target to be tracked in the predicted next frame image, the judgment of whether the target to be tracked is occluded is introduced, and different methods are used to obtain the final position of the target to be tracked in the predicted next frame image when the target to be tracked is not occluded or occluded, so the anti-interference rotating target tracking method of the present invention is used to obtain the final position of the target to be tracked in the predicted next frame image more accurately, and the accuracy of tracking the occluded and rotating targets is improved; and when the anti-interference rotating target tracking method of the present invention is used to obtain the final position of the target to be tracked in the predicted next frame image, the calculation process does not require a large amount of data training, the calculation process is simple, the hardware requirements are low, and it can be installed and run on embedded devices; In summary, the present invention solves the technical problem that most of the existing tracking methods cannot achieve accurate tracking when tracking occluded and rotating targets, and some can achieve relatively accurate tracking but the calculation process is too complicated and cannot be installed and run on embedded devices.

[0016] (2) In the anti-interference rotating target tracking method of the present invention, the n pairs of positive and negative symmetric angles set in step 2 are preferably selected from ±0.5 degrees, ±1 degrees, ±1.5 degrees and ±2 degrees. This setting can not only more comprehensively cover the rotation angles of the target to be tracked, but also simplify the calculation process, and the image affine rotation effect is obvious, thereby improving the tracking accuracy.

[0017] (3) In the anti-interference rotating target tracking method of the present invention, when obtaining the predicted rotation angle α of the target to be tracked in the second frame image in step 3 and obtaining the predicted rotation angle α of the target to be tracked in the next frame image in step 8, it is preferably also included a step of measuring the rotation angle α of the target to be tracked by using a scale-invariant feature transformation algorithm, so that the rotation angle α of the target to be tracked obtained in the end is more accurate, thereby improving the tracking accuracy.

[0018] (4) The anti-interference rotating target tracking method of the present invention introduces the judgment of whether the target to be tracked is occluded, and preferably provides two judgment criteria, which improves the anti-interference performance during the tracking process, and further improves the accuracy of tracking the occluded and rotating targets.

[0019] (5) The anti-interference rotating target tracking method of the present invention is used to track the rotating target. Experimental verification shows that the tracking accuracy can reach more than 96.7%. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of an embodiment of the anti-interference rotating target tracking method of the present invention; Figure 2 1 is a schematic diagram of the tracking effect on the corresponding frame image when the anti-interference rotating target tracking method embodiment of the present invention is used to track an aircraft; wherein: (a) is the first frame image; (b) is the 50th frame image; (c) is the 100th frame image; (d) is the 150th frame image; Figure 3 1 is a schematic diagram of the tracking effect on the corresponding frame image when the anti-interference rotating target tracking method embodiment of the present invention is used to track a ship; wherein: (e) is the first frame image; (f) is the 249th frame image; (g) is the 307th frame image; (h) is the 326th frame image; Figure 4 It is a comparative diagram of the accuracy curves of tracking a rotating target using different tracking methods. DETAILED DESCRIPTION

[0021] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] See also Figure 1 The present invention provides an anti-interference rotating target tracking method, comprising the following steps: Step 1: Obtain a first frame image, frame the position of the target to be tracked on the image, and obtain an image block of the target to be tracked on the first frame image; When the position of the target to be tracked is framed as above, manual framing can be used, or other methods such as introducing a lightweight deep learning neural network such as YOLO_X (a variant of the YOLO series (You Only Look Once) target detection model, proposed by the Alibaba Cloud team and Megvii Technology in 2021), YOLOv3 (the third version of the YOLO series (You Only Look Once) target detection model) and SSD (a single-stage target detection model) can be used to detect the target to be tracked. In this embodiment, manual framing is used; in order to improve the accuracy and stability of tracking, when the image block of the target to be tracked on the first frame image is obtained, the area of ​​the image block of the target to be tracked is preferably generally 2.5 to 3.5 times the area of ​​the target to be tracked, and its specific value is determined by the size of the target to be tracked, which is an empirical value; if the target to be tracked is an aircraft or a ship target, it is usually 3 times the size; if the target to be tracked is a vehicle target, it is usually 2.5 times the size; in this embodiment, the target to be tracked is an aircraft, and 3 times the size of the target area to be tracked is taken; Step 2: For the image blocks obtained in step 1, perform affine rotations of n pairs of positive and negative symmetric angles to obtain 2n rotated image blocks, where n is a natural number greater than 1; The absolute values ​​of the n pairs of positive and negative symmetric angles set above are preferably generally greater than zero and less than or equal to 3 degrees; in order to more comprehensively cover the angles of rotation between frames of the target to be tracked, and to simplify the calculation process, and to achieve an obvious image affine rotation effect, preferably the n is less than or equal to 4, and the n pairs of positive and negative symmetric angles set above are selected from ±0.5 degrees, ±1 degrees, ±1.5 degrees, and ±2 degrees; for example, when n is 2, the n pairs of positive and negative symmetric angles can be ±0.5 degrees and ±1 degrees; in this embodiment, in order to achieve higher tracking accuracy, preferably n is 4, that is, in this embodiment, the image blocks obtained in step 1 are subjected to affine rotations of ±0.5 degrees, ±1 degrees, ±1.5 degrees, and ±2 degrees, respectively, to obtain 8 rotated image blocks; Step 3: According to the image block obtained in step 1 and the 2n rotated image blocks obtained in step 2, the maximum target response map of the second frame image is obtained by using the kernel correlation filter model; then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2, the predicted rotation angle α of the target to be tracked in the second frame image is obtained; the rotation angle α of the target to be tracked refers to the rotation angle of the target to be tracked relative to the target to be tracked in the previous frame image; The specific process of obtaining the predicted rotation angle α of the target to be tracked in the second frame image can be as follows: the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 is directly used as the predicted rotation angle α of the target to be tracked in the second frame image; for example, the rotated image block corresponding to the maximum target response map of the second frame image is the image block obtained in step 1 in step 2 after performing an affine rotation of 0.5 degrees, then the predicted rotation angle α of the target to be tracked in the second frame image is obtained as 0.5 degrees; in addition to the direct acquisition described above, the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 can also be used as the predicted rotation angle of the first frame image obtained. The theoretical value of the rotation angle α of the target to be tracked in the two frames of images is obtained, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as the measured value of the rotation angle α of the target to be tracked in the second frame of the image to be predicted, and then the theoretical value and the measured value are averaged as the rotation angle α of the target to be tracked in the second frame of the image to be predicted; which method is used to obtain the rotation angle α of the target to be tracked in the second frame of the image to be predicted is determined according to the requirements for tracking accuracy in practical applications; in this embodiment, the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame of the image is selected from the affine rotation angles of the 2n rotated image blocks obtained in step 2, and is directly used as the rotation angle α of the target to be tracked in the second frame of the image to be predicted. In this embodiment, the rotation angle α of the target to be tracked in the second frame of the image to be predicted is 0.5 degrees; Step 4: taking the position corresponding to the maximum target response map of the second frame image obtained in step 3 as the predicted final position of the target to be tracked in the second frame image, and updating the parameters of the kernel correlation filter model in step 3 based on the final position; Step 5: Obtain the second frame image as the current frame image; Step 6: Based on the final position of the target to be tracked in the current frame image predicted by the previous frame image, an image block of the target to be tracked on the current frame image is obtained; When obtaining the image block of the target to be tracked on the current frame image, similar to step 1, the area of ​​the image block of the target to be tracked is preferably 2.5 to 3.5 times the area of ​​the target to be tracked, and the specific value is determined by the size of the target to be tracked and is an empirical value; if the target to be tracked is an aircraft or a ship target, it is usually 3 times the size; if the target to be tracked is a vehicle target, it is usually 2.5 times the size; in this embodiment, the target to be tracked is an aircraft, and it is 3 times the area of ​​the target to be tracked; Step 7: For the image block obtained in step 6, perform affine rotation of the angle α of the target to be tracked in the current frame image predicted based on the previous frame image, and the angle β in the nearest pool corresponding to the angle α, respectively, to obtain two rotated image blocks; the above-mentioned angle β in the nearest pool is the angle value closest to the angle α and whose absolute value is greater than the absolute value of the angle α among the n pairs of positive and negative symmetric angles set in step 2, except the angle α. When the angle α is equal to the angle with the largest absolute value among the n pairs of positive and negative symmetric angles set in step 2, the angle β in the nearest pool is the angle value closest to the angle α except the angle α; in this embodiment, the angle β in the nearest pool is the angle value closest to the angle α and whose absolute value is greater than the absolute value of the angle α among ±0.5 degrees, ±1 degrees, ±1.5 degrees and ±2 degrees except the angle α. When the angle α is equal to the angle with the largest absolute value among the n pairs of positive and negative symmetric angles set in step 2, the angle β in the nearest pool is the angle value closest to the angle α except the angle α; for example: when the angle α is equal to +1 degree, the angle β in the nearest pool is +1.5 degrees; when the angle α is equal to +2 degrees, the angle β in the nearest pool is +1.5 degrees; when the angle α is equal to -2 degrees, the angle β in the nearest pool is -1.5 degrees; when the current frame image is the second frame image, the predicted rotation angle α of the target to be tracked in the second frame image obtained in the above step 3 in this embodiment is 0.5 degrees, then the angle β in the nearest pool is 1 degree, and in this step 7, the image block of the target to be tracked on the second frame image obtained in step 6 is subjected to affine rotations of 0.5 degrees and 1 degree respectively to obtain two rotated image blocks; Step 8: According to the image block obtained in step 6 and the two rotated image blocks obtained in step 7, the maximum target response map of the next frame image is obtained by using the kernel correlation filter model whose parameters have been updated most recently; and then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles obtained when the two rotated image blocks are obtained in step 7, the predicted rotation angle α of the target to be tracked in the next frame image is obtained; The specific process of obtaining the predicted rotation angle α of the target to be tracked in the next frame image is the same as step 3. It can be: the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles of the two rotated image blocks obtained in step 7 is directly used as the predicted rotation angle α of the target to be tracked in the next frame image; or the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles of the two rotated image blocks obtained in step 7 is first used as the theoretical value of the rotation angle α of the target to be tracked in the next frame image, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as The measured value of the rotation angle α of the target to be tracked in the next frame image to be predicted is obtained, and then the theoretical value and the measured value are averaged as the rotation angle α of the target to be tracked in the next frame image to be predicted; which method is used to obtain the rotation angle α of the target to be tracked in the next frame image to be predicted is determined according to the requirements for tracking accuracy in practical applications; in this embodiment, the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles of the two rotated image blocks obtained in step 7 is directly used as the rotation angle α of the target to be tracked in the next frame image to be predicted; when the current frame image is the second frame image, the rotation angle α of the target to be tracked in the next frame image to be predicted (i.e., the third frame image) obtained in this embodiment is 1 degree; Step 9: According to the preset occlusion judgment standard of the target to be tracked, determine whether the target to be tracked is occluded; if not, execute step 10; if yes, execute step 11; In this embodiment, according to the preset occlusion judgment standard of the target to be tracked, the specific method for judging whether the target to be tracked is occluded is: judging according to the size relationship between the peak value of the maximum target response map of the next frame image obtained in step 8 and the set occlusion detection threshold: when the peak value of the maximum target response map of the next frame image obtained in step 8 is greater than the set occlusion detection threshold, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded; in order to improve the accuracy of the occlusion judgment of the target to be tracked, the above-mentioned set occlusion detection threshold is preferably 0.3-0.5, and its specific value is determined by the size and motion characteristics of the target to be tracked, and is an empirical value; usually when the target to be tracked is an aircraft target, the set occlusion detection threshold is 0.3; when When the target to be tracked is a ship target, the set occlusion detection threshold is 0.35; when the target to be tracked is a vehicle target or other targets except aircraft and ships, the set occlusion detection threshold is 0.4; in this embodiment, the target to be tracked is an aircraft, and the set occlusion detection threshold is 0.3; the above-mentioned specific method for judging whether the target to be tracked is occluded according to the preset occlusion judgment standard of the target to be tracked, in addition to the method of this embodiment, can also introduce lightweight deep learning neural networks such as YOLO_X, YOLOv3 and SSD to detect the target to be tracked, and judge according to the detection result: if the target to be tracked can be detected and the confidence is greater than 0.5, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded; Step 10: taking the position corresponding to the maximum target response map of the next frame image obtained in step 8 as the predicted final position of the target to be tracked in the next frame image, and based on the final position, updating the parameters of the kernel correlation filter model most recently used in step 8; then executing step 12; Step 11: Based on the maximum target response map of the next frame image obtained in step 8 and the motion displacement of the target to be tracked in the most recent M frames, the Kalman filter is integrated to obtain the predicted final position of the target to be tracked in the next frame image, where M is an integer greater than or equal to 3; if the total number of frames before the current frame image is less than M frames, the motion displacement of the target to be tracked from the first frame image to the current frame image is obtained by integrating the Kalman filter; In order to make the predicted final position of the target to be tracked in the next frame image more accurate when the target to be tracked is blocked, thereby achieving more accurate tracking, and to make the calculation process relatively simple, the above M is preferably usually 3 to 5, and its specific value is determined by the motion characteristics of the target to be tracked and is an empirical value; usually when the target to be tracked is an aircraft or a ship target, M is 3, and when the target to be tracked is a vehicle, M is 5; in this embodiment, M is 3; Step 12: Determine whether an end tracking instruction is received; if not, obtain the next frame image as the new current frame image, update the current frame image, and return to step 6; if yes, the tracking is completed.

[0023] In addition, the present invention also provides an embedded device, including a memory, a processor and a computer program stored in the memory; the processor executes the computer program to implement the steps of the above-mentioned anti-interference rotating target tracking method.

[0024] Figure 2 The figure is a schematic diagram of the tracking effect on the corresponding frame images when the anti-interference rotating target tracking method embodiment of the present invention is used to track an aircraft; wherein: (a) is the 1st frame image; (b) is the 50th frame image; (c) is the 100th frame image; and (d) is the 150th frame image. Figure 3 1 is a schematic diagram of the tracking effect on the corresponding frame images when the anti-interference rotating target tracking method embodiment of the present invention is used to track a ship; wherein: (e) is the 1st frame image; (f) is the 249th frame image; (g) is the 307th frame image; and (h) is the 326th frame image. Figure 2 and Figure 3 It can be seen that the anti-interference rotating target tracking method of the present invention can achieve accurate tracking of aircraft and ships, and can effectively track rotating targets, and estimate the target position when occlusion occurs, and keep tracking the target when the occlusion ends.

[0025] Figure 4 It is a comparative diagram of the accuracy curves of tracking a rotating target using different tracking methods. Figure 4 In the present invention, the tracking accuracies of BOOSTING (an integrated statistical learning tracking algorithm), MIL (multiple instance learning tracking algorithm), MEDIANFLOW (median optical flow tracking algorithm), Siamese R-CNN (a neural network tracking method for visual tracking by re-detection), ECO (efficient continuous convolution operator tracking algorithm), KCF (kernel correlation filtering algorithm), and RACFME (the anti-interference rotating target tracking method of the present invention) were compared, and the tracking accuracies were 47.6%, 42.1%, 8.2%, 90.5%, 77.8%, 73.4%, and 96.7%, respectively. It can be seen that among these methods, the anti-interference rotating target tracking method of the present invention has the highest tracking accuracy for rotating targets.

[0026] In summary, the anti-interference rotating target tracking method of the present invention improves the anti-interference performance during the tracking process and can accurately track occluded and rotating targets with high precision.

Claims

1. A method for tracking a rotating target with interference resistance, characterized in that: The following steps are involved: Step 1: Obtain a first frame image, frame the position of the target to be tracked on the image, and obtain an image block of the target to be tracked on the first frame image; Step 2: Performing affine rotations of n pairs of positive and negative symmetric angles on the image blocks obtained in step 1 to obtain 2n rotated image blocks, where n is a natural number greater than 1; Step 3: According to the image block obtained in step 1 and the 2n rotated image blocks obtained in step 2, a maximum target response map of the second frame image is obtained by using a kernel correlation filter model; Then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2, the predicted rotation angle α of the target to be tracked in the second frame image is obtained; the rotation angle α of the target to be tracked refers to the rotation angle of the target to be tracked relative to the target to be tracked in the previous frame image; Step 4: taking the position corresponding to the maximum target response map of the second frame image obtained in step 3 as the predicted final position of the target to be tracked in the second frame image, and updating the parameters of the kernel correlation filter model in step 3 based on the final position; Step 5: Obtain the second frame image as the current frame image; Step 6: based on the final position of the target to be tracked in the current frame image predicted by the previous frame image, an image block of the target to be tracked on the current frame image is obtained; Step 7: For the image block obtained in step 6, perform affine rotation of the angle α of the target to be tracked in the current frame image predicted based on the previous frame image, and the angle β in the nearest pool corresponding to the angle α, respectively, to obtain two rotated image blocks; the angle β in the nearest pool is the angle value closest to the angle α and having an absolute value greater than the absolute value of the angle α among the n pairs of positive and negative symmetric angles set in step 2, except the angle α. When the angle α is equal to the angle with the largest absolute value among the n pairs of positive and negative symmetric angles set in step 2, the angle β in the nearest pool is the angle value closest to the angle α except the angle α. Step 8: According to the image block obtained in step 6 and the two rotated image blocks obtained in step 7, the maximum target response map of the next frame image is obtained by using the kernel correlation filter model whose parameters have been updated most recently; and then, according to the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles when the two rotated image blocks are obtained in step 7, the predicted rotation angle α of the target to be tracked in the next frame image is obtained; Step 9: judging whether the target to be tracked is blocked according to the preset target to be tracked blockage judgment standard; If not, proceed to step 10; If yes, go to step 11; Step 10: taking the position corresponding to the maximum target response map of the next frame image obtained in step 8 as the predicted final position of the target to be tracked in the next frame image, and based on the final position, updating the parameters of the kernel correlation filter model most recently used in step 8; then executing step 12; Step 11: Based on the maximum target response map of the next frame image obtained in step 8 and the motion displacement of the target to be tracked in the most recent M frames, the Kalman filter is integrated to obtain the predicted final position of the target to be tracked in the next frame image, where M is an integer greater than or equal to 3; if the total number of frames before the current frame image is less than M frames, the motion displacement of the target to be tracked from the first frame image to the current frame image is obtained by integrating the Kalman filter; Step 12: Determine whether an end tracking instruction is received; if not, obtain the next frame image as a new current frame image, update the current frame image, and return to step 6; If yes, the tracking is completed.

2. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 1, the position of the target to be tracked is framed by manually framing the position of the target to be tracked or by introducing a lightweight deep learning neural network to detect the target to be tracked.

3. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 1 and step 6, the area of ​​the image block of the target to be tracked is 2.5 to 3.5 times the area of ​​the target to be tracked, and its specific value is determined by the size of the target to be tracked and is an empirical value; In step 2, the absolute values ​​of the set n pairs of positive and negative symmetry angles are all greater than zero and less than or equal to 3 degrees.

4. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 2, n is less than or equal to 4; the set n pairs of positive and negative symmetry angles are selected from ±0.5 degrees, ±1 degree, ±1.5 degrees and ±2 degrees.

5. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 3, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the second frame image is as follows: the affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 is directly used as the obtained predicted rotation angle α of the target to be tracked in the second frame image; In step 8, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the next frame image is: among the affine rotation angles of the two rotated image blocks obtained in step 7, the affine rotation angle of a rotated image block corresponding to the maximum target response map of the next frame image is directly used as the obtained predicted rotation angle α of the target to be tracked in the next frame image.

6. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 3, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the second frame image is: The affine rotation angle of a rotated image block corresponding to the maximum target response map of the second frame image among the affine rotation angles of the 2n rotated image blocks obtained in step 2 is used as the theoretical value of the rotation angle α of the target to be tracked in the predicted second frame image, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as the measurement value of the rotation angle α of the target to be tracked in the predicted second frame image, and then the theoretical value and the measured value are averaged to obtain the rotation angle α of the target to be tracked in the predicted second frame image; In step 8, the specific process of obtaining the predicted rotation angle α of the target to be tracked in the next frame image is: The affine rotation angle of the rotated image block corresponding to the maximum target response map of the next frame image among the affine rotation angles of the two rotated image blocks obtained in step 7 is used as the theoretical value of the rotation angle α of the target to be tracked in the predicted next frame image, and then the scale-invariant feature transformation algorithm is used to measure the rotation angle α of the target to be tracked, and the measurement result is used as the measured value of the rotation angle α of the target to be tracked in the predicted next frame image, and then the theoretical value and the measured value are averaged to obtain the rotation angle α of the target to be tracked in the predicted next frame image.

7. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 9, according to the preset occlusion judgment standard of the target to be tracked, the specific method for judging whether the target to be tracked is occluded is: The judgment is made according to the magnitude relationship between the peak value of the maximum target response map of the next frame image obtained in step 8 and the set occlusion detection threshold: when the peak value of the maximum target response map of the next frame image obtained in step 8 is greater than the set occlusion detection threshold, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded; The set occlusion detection threshold is 0.3-0.5, and its specific value is determined by the size and motion characteristics of the target to be tracked, and is an empirical value.

8. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 9, according to the preset occlusion judgment standard of the target to be tracked, the specific method for judging whether the target to be tracked is occluded is: A lightweight deep learning neural network is introduced to detect the target to be tracked, and a judgment is made based on the detection results: if the target to be tracked can be detected and the confidence level is greater than 0.5, it is judged as no, and the target to be tracked is not occluded; otherwise, it is judged as yes, and the target to be tracked is occluded.

9. The anti-interference rotating target tracking method according to claim 1, characterized in that: In step 11, M is 3-5, and its specific value is determined by the motion characteristics of the target to be tracked and is an empirical value.

10. An embedded 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 interference-resistant rotating target tracking method according to any one of claims 1 to 9.

Images