Target Tracking Method, Electronic Device, and Storage Medium

Abstract

The present application discloses a target tracking method, an electronic device, and a storage medium. Among them, the target tracking method includes: obtaining the latest tracking threshold of the tracked target based on the trajectory information of the tracked target in the target tracking list, and detecting the position information of each target object in the object set in the image to be measured; then analyzing whether the tracked target is included in the object set based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object; in response to the existence of a tracked target not included in the object set, taking the tracked target not included in the object set as a target to be processed, and determining whether to remove the trajectory information of the target to be processed from the target tracking list based on the trajectory information of the target to be processed and the latest loss threshold; updating the target tracking list based on the position information of each target object in the object set. The above solution can improve the accuracy of target tracking.

Description

Technical Field

[0001] This application relates to the field of computer vision technology, and particularly to an object tracking method, an electronic device, and a storage medium. Background Art

[0002] With the development of computer vision technology, multi-object tracking technology has been widely applied to scenarios such as security monitoring and driverless driving. Multi-object tracking technology can achieve motion perception of objects, and then master the motion state of the objects.

[0003] Currently, detection-based multi-object tracking is generally adopted. Specifically, the same objects in a time series of images are associated, and are distinguished by assigning IDs. However, since the object disappearance threshold is a fixed value, it is easy for objects to be mutually occluded or occluded by other objects. When an object is occluded, the object is considered to be in a lost state. When the occlusion ends and the object appears again, it will be regarded as a new object and tracked again. Therefore, it is difficult to ensure the accuracy of the object tracking result. In view of this, how to improve the accuracy of object tracking has become an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem to be solved by this application is to provide an object tracking method, an electronic device, and a storage medium, which can improve the accuracy of object tracking.

[0005] To solve the above technical problem, a first aspect of this application provides an object tracking method, including: obtaining the latest tracking threshold of the tracked objects based on the trajectory information of the tracked objects in the object tracking list; wherein, the latest tracking threshold includes the latest matching threshold and the latest disappearance threshold; detecting the position information of each target object in the object set in the to-be-detected image; analyzing whether the tracked objects are included in the object set based on the trajectory information of the tracked objects, the latest matching threshold, and the position information of the target objects; in response to there being tracked objects not included in the object set, taking the tracked objects not included in the object set as to-be-processed objects, and determining whether to remove the trajectory information of the to-be-processed objects from the object tracking list based on the trajectory information of the to-be-processed objects and the latest disappearance threshold; updating the object tracking list based on the position information of each target object in the object set.

[0006] To solve the above technical problem, a second aspect of this application provides an electronic device, including a memory and a processor coupled to each other. Program instructions are stored in the memory, and the processor is configured to execute the program instructions to implement the object tracking method in the first aspect above.

[0007] To solve the above technical problem, a third aspect of this application provides a computer-readable storage medium, storing program instructions that can be run by a processor, and the program instructions are used to implement the object tracking method in the first aspect above.

[0008] In the above solution, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained. The latest tracking thresholds include the latest matching threshold and the latest loss threshold, and the position information of each target object in the object set in the image to be measured is detected. Then, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked targets are included in the object set. In response to the existence of tracked targets that are not included in the object set, the tracked targets that are not included in the object set are used as the targets to be processed, and based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list. On this basis, based on the position information of each target object in the object set, the target tracking list is updated. On the one hand, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained, which helps to improve the accuracy of the tracking thresholds of the tracked targets. On the other hand, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked targets are included in the object set, thereby determining the correlation between the target objects in the image to be measured and the tracked targets. On this basis, based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list and update the target tracking list, which helps to improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0009] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings here are incorporated into the specification and constitute a part of this specification. These drawings show embodiments consistent with the present application and are used together with the specification to explain the technical solutions of the present application.

[0011] Figure 1 It is a schematic flowchart of an embodiment of the target tracking method of the present application;

[0012] Figure 2 It is a fitting schematic diagram of an embodiment of the trajectory information of the tracked targets;

[0013] Figure 3 It is a schematic diagram of an embodiment of the preset area to be measured;

[0014] Figure 4 It is a schematic flowchart of another embodiment of the target tracking method of the present application;

[0015] Figure 5 It is a schematic framework diagram of an embodiment of the target tracking device of the present application;

[0016] Figure 6 It is a schematic diagram of the framework of an embodiment of the electronic device of the present application;

[0017] Figure 7 It is a schematic diagram of the framework of an embodiment of the computer-readable storage medium of the present application. Specific embodiments

[0018] The following will combine the accompanying drawings of the specification to elaborate on the solutions of the embodiments of the present application in detail.

[0019] In the following description, specific details such as specific system architectures, interfaces, and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the present application.

[0020] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after. In addition, "multiple" in this article means two or more than two. In addition, the term "at least one" in this article represents any one of multiple or any combination of at least two of multiple. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set composed of A, B, and C. "Several" means at least one. The terms "first", "second", etc. in the specification, claims, and above-mentioned drawings of this article are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence.

[0021] Please refer to Figure 1 , Figure 1 It is a schematic flowchart of an embodiment of the target tracking method of the present application.

[0022] Specifically, the following steps can be included:

[0023] Step S11: Based on the trajectory information of the targets already tracked in the target tracking list, obtain the latest tracking threshold of the targets already tracked.

[0024] In an implementation scenario, the target tracking list may include the trajectory information of the tracked target, the ID of the tracked target, the type of the tracked target, the number of times the tracked target is lost, the number of image frames in which the tracked target is continuously detected, the confidence level of the tracked target, etc.; and the number of image frames in which the tracked target is continuously detected and the number of times the tracked target is lost will be updated according to the detection result of the image to be measured. In addition, the trajectory information of the tracked target is composed of the position information of the tracked target at different times, and the position information of the tracked target can be determined by the coordinates of the preset measurement area in the image to be measured, or can be determined by the circumscribed rectangle of the tracked target, and no specific limitation is made here.

[0025] In the embodiments of the present disclosure, the latest tracking thresholds include the latest matching threshold and the latest loss threshold, and the tracking thresholds are updated according to the trajectory information of the tracked target in the target tracking list.

[0026] In an implementation scenario, in order to obtain the latest tracking threshold of the tracked target, the motion direction of the tracked target can be analyzed based on the trajectory information of the tracked target, and the tracking threshold of the tracked target can be weighted based on the motion direction of the tracked target to obtain the latest tracking threshold of the tracked target. Specifically, based on the trajectory information of the tracked target, the latest position information of the tracked target can be selected. The latest position information of the tracked target is obtained by performing target detection on the preset measurement area in the latest captured image of the tracked target. Then, the position distance from the latest position information of the tracked target to the center position of the preset measurement area is obtained, and the position distance is compared with the preset position distance. When the position distance is less than the preset position distance, the position distance is positively correlated with the latest tracking threshold of the tracked target. When the position distance is not less than the preset position distance, the position distance is negatively correlated with the latest tracking threshold of the tracked target. The preset position distance can be set to 5 meters, 10 meters, etc., and the preset position distance can be determined according to the actual situation, and no specific limitation is made here.

[0027] In another implementation scenario, different from the foregoing implementation manner, it is possible to first count the number of image frames in which the tracked target has been continuously detected based on the trajectory information of the tracked target. It can be understood that the trajectory information of the tracked target consists of the continuous detection results of the to-be-tested images. However, during the tracking process, the tracked target may be lost. Therefore, the recent loss situation of the tracked target can be determined by counting the number of image frames in which the tracked target has been continuously detected. After obtaining the number of image frames in which the tracked target has been continuously detected, if the number of image frames does not meet the update condition, the latest tracking threshold of the tracked target can be determined as the preset tracking threshold; if the number of image frames meets the update condition, based on the trajectory information of the tracked target, the movement direction and movement speed of the tracked target can be analyzed, and based on the latest position information, movement direction and movement speed of the tracked target, the latest tracking threshold of the tracked target can be obtained. The update condition can be that the number of image frames is not less than the preset number of frame thresholds, and the preset number of frames can be set to 10 frames, 15 frames, etc.; the update condition can also be that the number of image frames is not less than the preset number of frame thresholds and the number of image frames has been updated. In the above manner, by counting the number of image frames in which the tracked target has been continuously detected based on the trajectory information of the tracked target, it is further determined whether the tracked target is lost, and then the tracking threshold of the tracked target is updated through the number of image frames, and the latest tracking threshold of the tracked target is obtained, which helps to improve the accuracy of the tracking threshold of the tracked target, and further improves the accuracy of target detection.

[0028] Further, based on the trajectory information of the tracked target, the movement direction and movement speed of the tracked target can be analyzed. Specifically, first, based on the trajectory information of the tracked target, a linear function is fitted. The linear function is the functional relationship between the displacement and time of the tracked target. The fitting method can be selected as the least squares method, the gradient descent method, etc., and no specific limitation is made here. Exemplarily, please refer to Figure 2 , Figure 2 is a fitting schematic diagram of an embodiment of the trajectory information of the tracked target. The functional relationship between the displacement and time of the tracked target can be obtained according to the trajectory information of the tracked target. Then, based on the linear function, the movement direction and movement speed of the tracked target are calculated. Specifically, the movement speed of the tracked target can be directly calculated according to the functional relationship, and the movement direction of the tracked target is determined. It is also possible to fit a motion trajectory fitting line according to the trajectory information of the tracked target. The slope of the motion trajectory fitting line is the movement speed of the tracked target, and the direction of the trajectory fitting line is the movement direction of the tracked target. In the above manner, a linear function is fitted through the trajectory information of the tracked target, and then the movement direction and movement speed of the tracked target are obtained through the linear function, which helps to improve the movement direction and movement speed of the tracked target, and further improves the accuracy of target tracking.

[0029] In addition, based on the preset area to be measured and the latest position information of the tracked target, the position weight of the tracked target can be obtained. The latest position information is obtained by performing target detection on the preset area to be measured in the latest captured image of the tracked target. Exemplarily, please refer to Figure 3 , Figure 3 is a schematic diagram of an embodiment of the preset area to be measured. The preset area to be measured can be the area of the captured image in the tracked list, that is, the area where the captured image is located in the figure is the preset area to be measured. Alternatively, a partial area of the captured image can be selected, that is, the circular area in the captured image in the figure is the preset area to be measured. The preset area to be measured can be determined according to the actual situation and is not specifically limited herein. Then, based on the preset area to be measured and the moving direction of the tracked target, the distance weight of the tracked target is obtained; based on the moving speed and the preset matching threshold, the updated matching threshold is obtained, and based on the moving speed, the distance weight, the position weight, and the preset loss threshold, the updated loss threshold is obtained. In the above manner, by determining the position weight and the distance weight of the tracked target based on the relevant information of the preset area to be measured and the tracked target, the updated matching threshold and the updated loss threshold of the tracked target can be determined by combining the position weight and the distance weight of the tracked target, further improving the accuracy of the updated matching threshold and the updated loss threshold of the tracked target, and thus improving the accuracy of target tracking.

[0030] In a specific implementation scenario, to obtain the position weight of the tracked target, the central position of the area to be measured can be obtained first, the distance from the tracked target to the central position can be obtained, and the position weight of the tracked target can be obtained based on the distance from the tracked target to the central position according to the mapping table. The mapping table can record the mapping relationship between the distance from the tracked target to the central position and the position weight. Exemplarily, when the distance from the tracked target to the central position is not less than 5 meters, the position weight is 0.8; when the distance from the tracked target to the central position is greater than 5 and not less than 10 meters, the position weight is 0.5; when the distance from the tracked target to the central position is greater than 10 and not less than 15 meters, the position weight is 0.2; when the distance from the tracked target to the central position is greater than 15 meters, the position weight is 0.

[0031] In another specific implementation scenario, as a possible implementation manner, the central position of the preset area to be measured can be obtained, the first distance from the tracked target to the central position can be obtained based on the latest position information of the tracked target, and then the position weight of the tracked target can be obtained by mapping the first distance, that is, mapping the first distance to between [0, 1], and taking the difference between 1 and the mapped value as the position weight of the tracked target. That is, the smaller the first distance, the larger the position weight value, that is, the position weight is negatively correlated with the first distance.

[0032] In yet another specific implementation scenario, different from the foregoing implementation manner, after obtaining the first distance from the tracked target to the center position, a Gaussian kernel function can be adopted, that is, with the center position of the preset area to be measured as the center point, monotonically decreasing and spreading around. The closer to the center point, the greater the weight. The specific expression is as follows:

[0033]

[0034] where w g represents the position weight, x represents the latest position of the tracked target, x` represents the center position, σ represents the width parameter of the function, which can be set as a fixed multiple of the width of the area to be measured, and ||x - x`|| 2 is the Euclidean distance between x and x'. It can be understood that the smaller the first distance, the greater the position weight. And when the first distance increases to a certain extent, the position weight decreases exponentially, so the position weight is negatively correlated with the first distance. By the above method, the position weight of the tracked target is determined through the first distance from the tracked target to the center position, which helps to increase the influence of the position information of the tracked target in the preset area to be measured on the target tracking effect, and further improves the accuracy of the position weight of the tracked target.

[0035] In a specific implementation scenario, in order to obtain the distance weight of the tracked target, based on the moving direction of the tracked target, determine the third distance between the tracked target and the boundary of the preset area to be measured in the moving direction, map the third distance to the range [0, 1], and then take the difference between 1 and the mapped value as the distance weight of the tracked target, that is, the smaller the third distance, the greater the distance weight value, that is, the distance weight is negatively correlated with the third distance.

[0036] In another specific implementation scenario, different from the foregoing implementation manner, first, based on the moving direction of the tracked target, determine the fitting line of the moving trajectory of the tracked target, then extend the fitting line of the moving trajectory, and take the intersection point of the fitting line of the moving trajectory and the boundary line of the preset area to be measured as the boundary position; based on the latest position information of the tracked target, obtain the second distance from the tracked target to the boundary position; based on the ratio of the second distance to the height of the preset area to be measured, determine the distance weight of the tracked target. The specific expression is as follows:

[0037]

[0038] where w disIt represents the distance weight of the tracked target, d represents the second distance, and H represents the height of the preset area to be measured. When the preset area to be measured is circular, H can represent the diameter of the preset area to be measured, and H can be determined according to the preset area to be measured, which is not specifically limited here. In the above manner, by fitting the movement trajectory line of the tracked target, the second distance from the tracked target to the boundary position is obtained, and based on the ratio of the second distance to the height of the preset area to be measured, the distance weight of the tracked target is determined, which helps to increase the influence of the position information of the tracked target on the target tracking effect, and further improves the accuracy of the distance weight of the tracked target.

[0039] In a specific implementation scenario, in order to obtain the updated matching threshold, the updated matching threshold can be determined based on the movement and the preset matching threshold. Specifically, when the movement speed is not less than the preset speed, the preset matching threshold is weighted based on the movement speed to obtain the determined updated matching threshold, that is, the greater the movement speed, the greater the weight. When the movement speed is less than the preset speed, half of the preset matching threshold is taken as the updated matching threshold.

[0040] In another specific implementation scenario, different from the foregoing implementation manner, the first matching threshold can be weighted by the product of the movement speed and the preset matching threshold, and the second matching threshold can be obtained by weighting based on the preset matching threshold. Then, the larger value of the first matching threshold and the second matching threshold is selected as the updated matching threshold. Exemplarily, the weight of the first matching threshold can be determined based on relevant parameters of the captured image, the area to be measured, etc. The specific expression is as follows:

[0041]

[0042] Among them, K match represents the updated matching threshold, v represents the movement speed, and k1 represents the preset matching threshold. is a control parameter, which can be determined according to the type of the tracked target or according to the actual situation. FPS is the frame rate of the captured video, and the captured image is the frame image of the captured video. h represents the height of the preset area to be measured. When the preset area to be measured is circular, h can represent the diameter of the preset area to be measured. h can be determined according to the preset area to be measured, and no specific limitation is made here. Specifically, the product of the control parameter and the frame rate of the captured video can be obtained first, and then the ratio of twice the product to the height of the preset area to be measured is taken as the weight value of the product of the moving speed and the preset matching threshold, that is, the first matching threshold is obtained; then the preset matching threshold is weighted with a weight of 0.5 to obtain the second matching threshold. Through the above method, by respectively determining the first matching threshold and the second matching threshold based on the moving speed and the preset matching threshold, and then selecting one of them as the updated matching threshold, the accuracy of the updated matching threshold is improved, and the relevance between the updated matching threshold and the tracked target is increased.

[0043] In a specific implementation scenario, in order to obtain the updated loss threshold, the larger one of the distance weight and the position weight can be selected, and the selected weight value, the preset loss threshold and the moving speed are used as the first loss threshold; then the preset loss threshold is weighted, and the weight is 0.5, to obtain the second loss threshold. On this basis, the larger one of the first loss threshold and the second loss threshold is selected as the updated loss threshold.

[0044] In another specific implementation scenario, different from the foregoing implementation manner, the target weight can be obtained by fusing the distance weight and the position weight. The fusion method can be accumulation, or can be weighted first and then accumulated, or can be multiplied. After obtaining the target weight, the product of the moving speed, the target weight and the preset loss threshold is used as the first loss threshold, and the preset loss threshold is weighted, and the weight can be set to 0.5, to obtain the second loss threshold; the larger one of the first loss threshold and the second loss threshold is selected as the updated loss threshold. The specific expression is as follows:

[0045]

[0046] Among them, K loss represents the updated loss threshold, β(w g + dis )·k2 represents the first loss threshold, represents the second loss threshold, w g represents the position weight, w disLet $\alpha$ represent the distance weight, $\beta$ represent the control parameter which is determined by the moving speed, and $k_2$ represent the preset loss threshold. In the above method, the first loss threshold and the second loss threshold are respectively determined by the moving speed, the distance weight, the position weight, and the preset loss threshold, and then one of them is selected as the updated loss threshold, which improves the accuracy of the updated loss threshold and increases the correlation between the updated loss threshold and the tracked target.

[0047] Step S12: Detect the position information of each target object in the object set in the image to be measured.

[0048] In an implementation scenario, the image to be measured can be a video frame image selected from video images or discrete captured images. The image to be measured can be determined according to the actual situation and is not specifically limited here.

[0049] In an implementation scenario, the object set includes at least one target object and the position information of the target object, and may also include the type of the target object, the attributes of the target object, etc. In addition, the position information of the target object can be determined by the bounding rectangle of the target object or by the coordinate position of the target object in the image to be measured. The position information of the target object can be determined according to the actual situation and is not specifically limited here.

[0050] In an implementation scenario, the image to be measured can be detected by a target detection network, and the target detection network can include but is not limited to CNN (Convolution Neural Network), RNN (Recurrent Neural Network), etc.

[0051] Step S13: Analyze whether the tracked target is included in the object set based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object.

[0052] In an implementation scenario, the target position information of the tracked target can be obtained based on the trajectory information of the tracked target. Specifically, it is determined whether the latest position information of the tracked target in the target tracking list is obtained from the previous detection. If it is obtained from the previous detection, the latest position information of the tracked target is used as the target position information of the tracked target. If it is not obtained from the previous detection, it is further determined whether the tracking threshold of the tracked target is updated. If it is updated, prediction is performed based on a linear function to obtain the target position information of the tracked target. If it is not updated, the position information of the tracked target obtained from the most recent detection in the target tracking list is used as the target position information of the tracked target. Further, based on the target position information of the tracked target and the position information of the target object, the position distance between the tracked target and the target object is obtained, and based on the magnitude relationship between the position distance and the latest matching threshold, it is determined whether the tracked target is included in the object set.

[0053] In another implementation scenario, different from the foregoing implementation manner, the latest position information of the tracked target can be obtained based on the trajectory information of the tracked target. The latest position information is obtained by performing target detection on a preset area to be measured in the latest captured image of the tracked target, and then based on the latest position information of the tracked target and the position information of the target object, the position distance between the tracked target and the target object is obtained. On this basis, based on the position distance and the latest matching threshold, it is determined whether the tracked target is included in the object set. By the above method, the position distance is obtained through the latest position information of the tracked target and the position information of the target object, and based on the position distance and the latest matching threshold, it is determined whether the tracked target is included in the object set, thereby improving the accuracy of matching between the tracked target and the object set while improving the accuracy of target tracking.

[0054] In a specific implementation scenario, it can be determined that the tracked target is included in the object set in response to the position distance being less than the latest matching threshold; and it can be determined that the tracked target is not included in the object set in response to the position distance being not less than the latest matching threshold. By the above method, by comparing the position distance with the latest matching threshold, it is determined whether the tracked target is included in the object set, which helps to improve the accuracy of matching between the tracked target and the target object in the object set.

[0055] In an implementation scenario, in response to there being a tracked target included in the object set, the number of continuously detected image frames of the tracked target is updated, that is, on the basis of the original number of image frames, it is incremented once to obtain the updated number of continuously detected image frames.

[0056] Step S14: In response to the existence of a tracked target not included in the object set, use the tracked target not included in the object set as the target to be processed, and based on the trajectory information of the target to be processed and the latest loss threshold, determine whether to remove the trajectory information of the target to be processed from the target tracking list.

[0057] In an implementation scenario, before determining whether to remove the trajectory information of the target to be processed from the target tracking list based on the trajectory information of the target to be processed and the latest loss threshold, update the loss count of the target to be processed, that is, add 1 to the original loss count of the target to be processed to obtain the loss count. Further, based on the magnitude relationship between the loss count of the target to be processed and the latest loss threshold, determine whether to remove the trajectory information of the target to be processed from the target tracking list; in response to the loss count of the target to be processed being not less than the latest loss threshold, remove the trajectory information of the target to be processed from the target tracking list; in response to the loss count of the target to be processed being less than the latest loss threshold, continue to detect the next tracked target. In the above manner, before determining whether to remove the trajectory information of the target to be processed from the target tracking list, the loss count is first updated, improving the accuracy of the loss count of the tracked target, and then making a comparison to determine whether the tracked target is lost, further improving the accuracy of target tracking.

[0058] Step S15: Update the target tracking list based on the position information of each target object within the object set.

[0059] In an implementation scenario, update the target tracking list based on the position information of each target object within the object set. Specifically, in the case where the tracked target is included in the object set, update the newly detected position information of the tracked target to the target tracking list; in the case where the tracked target is not included in the object set, update the position information of the new target object within the object set to the target tracking list, update the type of the new target object to the target tracking list, and assign a new ID to the new target object.

[0060] In the above solution, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained. The latest tracking thresholds include the latest matching threshold and the latest loss threshold, and the position information of each target object in the object set in the image to be measured is detected. Then, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target objects, it is analyzed whether the tracked targets are included in the object set. In response to the existence of tracked targets not included in the object set, the tracked targets not included in the object set are used as the targets to be processed, and based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list. On this basis, based on the position information of each target object in the object set, the target tracking list is updated. On the one hand, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained, which helps to improve the accuracy of the tracking thresholds of the tracked targets. On the other hand, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target objects, it is analyzed whether the tracked targets are included in the object set, thereby determining the correlation between the target objects in the image to be measured and the tracked targets. On this basis, based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list and update the target tracking list, which helps to improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0061] Please refer to Figure 4 , Figure 4 which is a schematic flowchart of another embodiment of the target tracking method of the present application.

[0062] Specifically, the following steps may be included:

[0063] Step S401: Based on the trajectory information of the tracked targets in the target tracking list, obtain the latest tracking thresholds of the tracked targets.

[0064] Specifically, the latest tracking thresholds include the latest matching threshold and the latest loss threshold. The method of obtaining the latest tracking thresholds of the tracked targets based on the trajectory information of the tracked targets in the target tracking list may refer to the method in the foregoing disclosed embodiment and will not be elaborated herein.

[0065] Step S402: Based on the trajectory information of the tracked targets in the target tracking list, determine the latest position information of the updated targets and the latest position information of the lost targets.

[0066] In an implementation scenario, based on the trajectory information of the tracked targets in the target tracking list, the tracked targets included in the previous detection are determined. The tracked targets included in the previous detection are used as updated targets, and the tracked targets not included in the previous detection are used as lost targets. Furthermore, the latest position information of the updated targets and the latest position information of the lost targets are determined. The determination method of the latest position information can refer to the method in the foregoing disclosed embodiments and will not be elaborated here.

[0067] Step S403: Detect the position information of each target object in the object set in the to-be-detected image.

[0068] Specifically, the detection method can refer to the detection method in the foregoing disclosed embodiments and will not be elaborated here.

[0069] Step S404: Determine the first position distance between the updated target and the target object, and determine the second position distance between the lost target and the target object.

[0070] Specifically, the determination method of the first position distance and the second position distance can refer to the determination method of the position distance in the foregoing disclosed embodiments and will not be elaborated here.

[0071] Step S405: Whether the first position distance is less than the latest matching threshold; if not, execute step S406; otherwise, execute step S407.

[0072] In an implementation scenario, matching can be performed through a greedy algorithm, a Hungarian algorithm, etc., that is, determine whether the first position distance is less than the latest matching threshold, and update the first target matching list and the first target non-matching list. It should be noted that the first target matching list and the first target non-matching list include the tracked targets.

[0073] Step S406: Update the first target non-matching list.

[0074] In an implementation scenario, when the first position distance is not less than the latest matching threshold, add the updated target to the first target non-matching list.

[0075] Step S407: Update the first target matching list.

[0076] In an implementation scenario, when the first position distance is less than the latest matching threshold, add the updated target to the first target matching list.

[0077] Step S408: Whether the second position distance is less than the latest matching threshold; if not, execute step S409; otherwise, execute step S410.

[0078] In an implementation scenario, matching can be performed through algorithms such as the greedy algorithm and the Hungarian algorithm, that is, determining whether the second position distance is less than the latest matching threshold, and updating the second target matching list and the second target unmatched list. It should be noted that the second target matching list and the second target unmatched list contain the tracked targets.

[0079] Step S409: Update the second non-target matching list.

[0080] In an implementation scenario, when the second position distance is not less than the latest matching threshold, add the updated target to the second target unmatched list.

[0081] Step S410: Update the second target matching list.

[0082] In an implementation scenario, when the second position distance is less than the latest matching threshold, add the updated target to the second target matching list.

[0083] Step S411: Perform re-matching and trajectory verification.

[0084] In an implementation scenario, since the updated targets and the lost targets are matched separately, it is possible that a target object is matched with a target in the first target matching list and also with a target in the second target matching list. Therefore, re-matching and trajectory verification can effectively remove duplicate matching situations, that is, remove the duplicate-matched tracked targets in the first target matching list or the second target matching list.

[0085] Step S412: Obtain the target unmatched list and the target matching list.

[0086] In an implementation scenario, re-matching and trajectory verification effectively remove the duplicate-matched tracked targets in the first target matching list or the second target matching list, add the screened duplicate-matched tracked targets to the target unmatched list, and update the information in the first target unmatched list and the second target unmatched list to the target unmatched list; and update the information in the first target matching list and the second target matching list to the target matching list.

[0087] Furthermore, after determining the target matching list, match the tracked targets in the target matching list with the target objects in the object set to obtain the unmatched target objects in the object set, that is, the new target objects. Assign new IDs to the new target objects, and update the IDs, position information, types, confidence levels, etc. of the new target objects to the target tracking list. In addition, set the number of consecutive detected image frames of the new target objects to 1.

[0088] Step S413: Based on the target matching list, update the number of image frames and trajectory information of the tracked targets.

[0089] In one implementation scenario, based on the target matching list, the number of consecutive detected image frames and the trajectory information of the tracked targets in the target matching list are updated. Specifically, the number of consecutive detected image frames of the tracked targets in the target matching list is incremented once, and the position information of the tracked targets in the object set in the to-be-detected image is updated to the target tracking list, thereby obtaining the updated trajectory information of the tracked target objects.

[0090] Step S414: Update the loss count of the tracked targets based on the target unmatched list.

[0091] In one implementation scenario, based on the tracked targets in the target unmatched list, the loss count of the tracked targets is updated, and the tracked targets in the target unmatched list are used as the targets to be processed.

[0092] Step S415: Whether the loss count of the targets to be processed in the target unmatched list is not less than the latest loss threshold; if so, execute Step S416; otherwise, execute Step S417.

[0093] Step S416: Remove the trajectory information of the targets to be processed from the target tracking list.

[0094] Specifically, in response to the loss count of the targets to be processed being not less than the latest loss threshold, the trajectory information of the targets to be processed is removed from the target tracking list.

[0095] Step S417: End.

[0096] Specifically, it is possible to analyze one by one whether the tracked targets in the target tracking list are included in the object set, and further determine whether to remove the trajectory information of the targets to be processed from the target tracking list until all the tracked targets in the target tracking list are detected. Further, when it is detected here that there is an input of a to-be-detected image, continue to execute the steps of obtaining the latest tracking threshold of the tracked targets based on the trajectory information of the tracked targets in the target tracking list to update the target tracking list.

[0097] In the above solution, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained. The latest tracking thresholds include the latest matching threshold and the latest loss threshold, and the position information of each target object in the object set in the image to be measured is detected. Then, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked targets are included in the object set. In response to the existence of tracked targets not included in the object set, the tracked targets not included in the object set are used as the targets to be processed, and based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list. On this basis, based on the position information of each target object in the object set, the target tracking list is updated. On the one hand, based on the trajectory information of the tracked targets in the target tracking list, the latest tracking thresholds of the tracked targets are obtained, which helps to improve the accuracy of the tracking thresholds of the tracked targets. On the other hand, based on the trajectory information of the tracked targets, the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked targets are included in the object set, thereby determining the correlation between the target objects in the image to be measured and the tracked targets. On this basis, based on the trajectory information of the targets to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the targets to be processed from the target tracking list and update the target tracking list, which helps to improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0098] Those skilled in the art can understand that in the above method of the specific implementation manner, the writing order of each step does not mean a strict execution order that constitutes any limitation to the implementation process, and the specific execution order of each step should be determined according to its function and possible internal logic.

[0099] Please refer to Figure 5 , Figure 5It is a schematic framework diagram of an embodiment of the target tracking device of the present application. The target tracking device 50 includes an acquisition module 51, a detection module 52, an analysis module 53, a determination module 54, and an update module 55. Among them, the acquisition module 51 is used to obtain the latest tracking threshold of the tracked target based on the trajectory information of the tracked target in the target tracking list. The latest tracking threshold includes the latest matching threshold and the latest loss threshold. The detection module 52 is used to detect the position information of each target object in the object set in the image to be measured. The analysis module 53 is used to analyze whether the tracked target is included in the object set based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object. The determination module 54 is used to, in response to the existence of a tracked target that is not included in the object set, use the tracked target that is not included in the object set as the target to be processed, and determine whether to remove the trajectory information of the target to be processed from the target tracking list based on the trajectory information of the target to be processed and the latest loss threshold. The update module 55 is used to update the target tracking list based on the position information of each target object in the object set.

[0100] In the above solution, on the one hand, based on the trajectory information of the tracked target in the target tracking list, the latest tracking threshold of the tracked target is obtained, which helps to improve the accuracy of the tracking threshold of the tracked target. On the other hand, based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked target is included in the object set, thereby determining the correlation between the target object in the image to be measured and the tracked target. On this basis, based on the trajectory information of the target to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the target to be processed from the target tracking list and update the target tracking list, which helps to improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0101] In some disclosed embodiments, the acquisition module 51 includes a statistics sub-module and an analysis sub-module. The statistics sub-module is used to count the number of image frames in which the tracked target has been continuously detected based on the trajectory information of the tracked target. The analysis sub-module is used to, in response to the number of image frames meeting the update condition, analyze the movement direction and movement speed of the tracked target based on the trajectory information of the tracked target, and obtain the latest tracking threshold of the tracked target based on the latest position information, movement direction, and movement speed of the tracked target.

[0102] Therefore, by counting the number of image frames in which the tracked target has been continuously detected based on the trajectory information of the tracked target, it is determined whether the tracked target is lost, and then through the number of image frames, it is determined to update the tracking threshold of the tracked target to obtain the latest tracking threshold of the tracked target, which helps to improve the accuracy of the tracking threshold of the tracked target, and further improves the accuracy of target detection.

[0103] In some disclosed embodiments, the update condition includes: the number of image frames is not less than a preset frame number threshold; the obtaining module 51 further includes a determination sub-module, and the determination sub-module is configured to determine that the latest tracking threshold of the tracked target is the preset tracking threshold in response to the number of image frames not meeting the update condition.

[0104] In some disclosed embodiments, the latest position information is obtained by performing target detection on a preset area to be measured in the latest captured image of the tracked target. The analysis sub-module includes a first determination unit, a second determination unit, a third determination unit, and a fourth determination unit. The first determination unit is configured to obtain the position weight of the tracked target based on the preset area to be measured and the latest position information of the tracked target. The second determination unit is configured to obtain the distance weight of the tracked target based on the preset area to be measured and the moving direction of the tracked target. The third determination unit is configured to obtain an updated matching threshold based on the moving speed and the preset matching threshold. The fourth determination unit is configured to obtain an updated loss threshold based on the moving speed, the distance weight, the position weight, and the preset loss threshold.

[0105] Therefore, by determining the position weight and the distance weight of the tracked target based on the preset area to be measured and the relevant information of the tracked target, and then combining the position weight and the distance weight of the tracked target, the updated matching threshold and the updated loss threshold of the tracked target can be determined, further improving the accuracy of the updated matching threshold and the updated loss threshold of the tracked target, and thus improving the accuracy of target tracking.

[0106] In some disclosed embodiments, the third determination unit includes a weighting sub-unit and a selection sub-unit. The weighting sub-unit is configured to weight and obtain a first matching threshold based on the product of the moving speed and the preset matching threshold, and perform weighting processing based on the preset matching threshold to obtain a second matching threshold. The selection sub-unit is configured to select the larger value of the first matching threshold and the second matching threshold as the updated matching threshold.

[0107] Therefore, by respectively determining the first matching threshold and the second matching threshold based on the moving speed and the preset matching threshold, and then selecting one of them as the updated matching threshold, the accuracy of the updated matching threshold is improved, and the relevance between the updated matching threshold and the tracked target is increased.

[0108] In some disclosed embodiments, the fourth determination unit includes a fusion sub-unit, a determination sub-unit, and a selection sub-unit. Among them, the fusion sub-unit is configured to fuse and obtain a target weight based on the distance weight and the position weight. The determination sub-unit is configured to use the product of the moving speed, the target weight, and the preset loss threshold as the first loss threshold, and perform weighting processing based on the preset loss threshold to obtain a second loss threshold. The selection sub-unit is configured to select the larger value of the first loss threshold and the second loss threshold as the updated loss threshold.

[0109] Therefore, by using the movement speed, distance weight, position weight, and preset loss threshold, the first loss threshold and the second loss threshold are respectively determined, and then one of them is selected as the updated loss threshold, which improves the accuracy of the updated loss threshold and increases the correlation between the updated loss threshold and the tracked target.

[0110] In some disclosed embodiments, the first determination unit includes an acquisition subunit, a calculation subunit, and a determination subunit. Among them, the acquisition subunit is used to acquire the central position of the preset area to be measured; the calculation subunit is used to obtain the first distance from the tracked target to the central position based on the latest position information of the tracked target; the determination subunit is used to determine the position weight of the tracked target based on the first distance; wherein, the position weight is negatively correlated with the first distance.

[0111] Therefore, by using the first distance from the tracked target to the central position to determine the position weight of the tracked target, it helps to increase the influence of the position information of the tracked target in the preset area to be measured on the target tracking effect, and further improves the accuracy of the position weight of the tracked target.

[0112] In some disclosed embodiments, the second determination unit includes a fitting subunit, a selection subunit, a calculation subunit, and a determination subunit. Among them, the fitting subunit is used to determine the motion trajectory fitting line of the tracked target based on the motion direction of the tracked target; the selection subunit is used to extend the motion trajectory fitting line and use the intersection point of the motion trajectory fitting line and the boundary line of the preset area to be measured as the boundary position; the calculation subunit is used to obtain the second distance from the tracked target to the boundary position based on the latest position information of the tracked target; the determination subunit is used to determine the distance weight of the tracked target based on the ratio of the second distance to the height of the preset area to be measured.

[0113] Therefore, by using the motion trajectory fitting line of the tracked target to obtain the second distance from the tracked target to the boundary position and determining the distance weight of the tracked target based on the ratio of the second distance to the height of the preset area to be measured, it helps to increase the influence of the position information of the tracked target on the target tracking effect, and further improves the accuracy of the distance weight of the tracked target.

[0114] In some disclosed embodiments, the analysis sub-module includes a fitting unit and a calculation unit. Among them, the fitting unit is used to fit a linear function based on the trajectory information of the tracked target; wherein, the linear function is the functional relationship between the displacement and time of the tracked target; the calculation unit is used to calculate the motion direction and motion speed of the tracked target based on the linear function.

[0115] Therefore, by using the trajectory information of the tracked target, a linear function is fitted, and then, through the linear function, the moving direction and moving speed of the tracked target are obtained, which helps to improve the moving direction and moving speed of the tracked target, and further improves the accuracy of target tracking.

[0116] In some disclosed embodiments, the analysis module 53 includes a determination sub-module, a calculation sub-module, and a judgment sub-module. Among them, the determination sub-module is used to obtain the latest position information of the tracked target based on the trajectory information of the tracked target; the calculation sub-module is used to obtain the position distance between the tracked target and the target object based on the latest position information of the tracked target and the position information of the target object; the judgment sub-module is used to determine whether the tracked target is included in the object set based on the position distance and the latest matching threshold.

[0117] Therefore, by using the latest position information of the tracked target and the position information of the target object, the position distance is obtained, and based on the position distance and the latest matching threshold, it is determined whether the tracked target is included in the object set, thereby improving the accuracy of matching between the tracked target and the target object in the object set while improving the accuracy of target tracking.

[0118] In some disclosed embodiments, the judgment sub-module includes a first determination unit and a second determination unit. The first determination unit is used to determine that the tracked target is included in the object set in response to the position distance being less than the latest matching threshold; the second determination unit is used to determine that the tracked target is not included in the object set in response to the position distance being not less than the latest matching threshold.

[0119] Therefore, by comparing the position distance with the latest matching threshold to determine whether the tracked target is included in the object set, it helps to improve the accuracy of matching between the tracked target and the target object in the object set.

[0120] In some disclosed embodiments, the update module 55 includes an update sub-module, a determination sub-module, and a judgment sub-module. Among them, the update sub-module is used to update the loss times of the target to be processed; the determination sub-module is used to determine whether to remove the trajectory information of the target to be processed from the target tracking list based on the magnitude relationship between the loss times of the target to be processed and the latest loss threshold; the judgment sub-module is used to remove the trajectory information of the target to be processed from the target tracking list in response to the loss times of the target to be processed being not less than the latest loss threshold.

[0121] Therefore, before determining whether to remove the trajectory information of the target to be processed from the target tracking list, the loss times are updated first, which improves the accuracy of the loss times of the tracked target, and then a comparison is made to determine whether the tracked target is lost, further improving the accuracy of target tracking.

[0122] Please refer to Figure 6 ,Figure 6 : is a schematic diagram of a framework of an embodiment of an electronic device of the present application. The electronic device 60 includes a memory 61 and a processor 62 coupled to each other, the memory 61 stores program instructions, and the processor 62 is used to execute the program instructions to implement the steps in any of the above target tracking method embodiments. Specifically, the electronic device 60 may include but is not limited to: a desktop computer, a laptop computer, a server, a mobile phone, a tablet computer, etc., which are not limited here.

[0123] Specifically, the processor 62 is used to control itself and the memory 61 to implement the steps in any of the above-mentioned target tracking method embodiments. The processor 62 can also be called a CPU (Central Processing Unit). The processor 62 may be an integrated circuit chip with signal processing capabilities. The processor 62 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 62 can be implemented by an integrated circuit chip.

[0124] In the above scheme, the electronic device 60 can implement the steps in any of the above target tracking method embodiments. On the one hand, based on the trajectory information of the tracked target in the target tracking list, the latest tracking threshold of the tracked target is obtained, which helps to improve the accuracy of the tracking threshold of the tracked target. On the other hand, based on the trajectory information of the tracked target and the latest matching threshold, and the position information of the target object, it is analyzed whether the tracked target is included in the object set, and then the correlation between the target object in the image to be tested and the tracked target is determined. On this basis, based on the trajectory information of the target to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the target to be processed from the target tracking list, and the target tracking list is updated, which helps to improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0125] See also Figure 7 , Figure 7 The computer-readable storage medium 70 stores program instructions 71 that can be executed by a processor, and the program instructions 71 are used to implement the steps in any of the above target tracking method embodiments.

[0126] In the above solution, the computer-readable storage medium 70 can implement the steps in any of the above-described target tracking method embodiments. On the one hand, based on the trajectory information of the targets being tracked in the target tracking list, the latest tracking threshold for the targets being tracked is obtained, which helps improve the accuracy of the tracking threshold for the targets being tracked. On the other hand, based on the trajectory information of the targets being tracked, the latest matching threshold, and the position information of the target object, it is analyzed whether the targets being tracked are included in the object set, thereby determining the correlation between the target object and the targets being tracked in the image to be measured. On this basis, based on the trajectory information of the target to be processed and the latest loss threshold, it is determined whether to remove the trajectory information of the target to be processed from the target tracking list and update the target tracking list, which helps improve the accuracy of the target tracking list. Therefore, the accuracy of target tracking can be improved.

[0127] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the methods described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0128] The above descriptions of the various embodiments tend to emphasize the differences between the various embodiments. The same or similar parts can be referred to each other. For the sake of brevity, they will not be repeated in this article.

[0129] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0130] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0131] In addition, the functional units in the various embodiments of the present application can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0132] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0133] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A target tracking method, characterized in that, Including: Based on the trajectory information of the tracked targets in the target tracking list, obtaining the latest tracking thresholds for the tracked targets; wherein, the latest tracking thresholds include the latest matching threshold and the latest loss threshold; Detecting the position information of each target object in the object set in the to-be-detected image; Based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object, analyzing whether the tracked target is included in the object set; In response to the existence of a tracked target not included in the object set, using the tracked target not included in the object set as a target to be processed, and based on the trajectory information of the target to be processed and the latest loss threshold, determining whether to remove the trajectory information of the target to be processed from the target tracking list; Updating the target tracking list based on the position information of each target object in the object set.

2. The method according to claim 1, characterized in that, The obtaining the latest tracking thresholds for the tracked targets based on the trajectory information of the tracked targets in the target tracking list includes: Based on the trajectory information of the tracked target, counting the number of image frames in which the tracked target has been continuously detected; In response to the number of image frames satisfying the update condition, based on the trajectory information of the tracked target, analyzing to obtain the movement direction and movement speed of the tracked target, and based on the latest position information, movement direction, and movement speed of the tracked target, obtaining the latest tracking thresholds for the tracked target.

3. The method according to claim 2, characterized in that, The update condition includes: the number of image frames is not less than a preset number-of-frames threshold; And / or, the method further includes: in response to the number of image frames not satisfying the update condition, determining the latest tracking thresholds for the tracked target as the preset tracking thresholds.

4. The method according to claim 2, characterized in that The latest position information is obtained by performing target detection on a preset to-be-detected area in the latest captured image of the tracked target. The obtaining the latest tracking thresholds for the tracked target based on the latest position information, movement direction, and movement speed of the tracked target includes: Based on the preset to-be-detected area and the latest position information of the tracked target, obtaining the position weight of the tracked target, and based on the preset to-be-detected area and the movement direction of the tracked target, obtaining the distance weight of the tracked target; Based on the movement speed and the preset matching threshold, obtaining an updated matching threshold, and based on the movement speed, the distance weight, the position weight, and the preset loss threshold, obtaining an updated loss threshold.

5. The method according to claim 4, wherein The obtaining an updated matching threshold based on the movement speed and the preset matching threshold includes: Based on the product of the movement speed and the preset matching threshold, obtaining a first matching threshold through weighted processing, and based on the preset matching threshold for weighted processing, obtaining a second matching threshold; Selecting the larger value of the first matching threshold and the second matching threshold as the updated matching threshold.

6. The method according to claim 4, characterized in that The obtaining an updated loss threshold based on the movement speed, the distance weight, the position weight, and the preset loss threshold includes: Based on the distance weight and the position weight, fusing to obtain a target weight; Multiply the motion speed, the target weight, and the preset loss threshold to obtain a first loss threshold, and perform weighted processing based on the preset loss threshold to obtain a second loss threshold; Select the larger value of the first loss threshold and the second loss threshold as the updated loss threshold.

7. The method according to claim 4, characterized in that, The obtaining the position weight of the tracked target based on the preset measurement area and the latest position information of the tracked target includes: Obtain the central position of the preset measurement area; Based on the latest position information of the tracked target, obtain a first distance from the tracked target to the central position; Based on the first distance, determine the position weight of the tracked target; wherein, the position weight is negatively correlated with the first distance.

8. The method according to claim 4, wherein The obtaining the distance weight of the tracked target based on the preset measurement area and the moving direction of the tracked target includes: Based on the moving direction of the tracked target, determine the motion trajectory fitting line of the tracked target; Extend the motion trajectory fitting line, and use the intersection point of the motion trajectory fitting line and the boundary line of the preset measurement area as the boundary position; Based on the latest position information of the tracked target, obtain a second distance from the tracked target to the boundary position; Based on the ratio of the second distance to the height of the preset measurement area, determine the distance weight of the tracked target.

9. The method according to claim 1, wherein The analyzing whether the tracked target is included in the object set based on the trajectory information of the tracked target, the latest matching threshold, and the position information of the target object includes: Based on the trajectory information of the tracked target, obtain the latest position information of the tracked target; Based on the latest position information of the tracked target and the position information of the target object, obtain the position distance between the tracked target and the target object; Based on the magnitude relationship between the position distance and the latest matching threshold, determine whether the tracked target is included in the object set.

10. The method according to claim 1, wherein Before determining whether to remove the trajectory information of the target to be processed from the target tracking list based on the trajectory information of the target to be processed and the latest loss threshold, the method further includes: Update the loss count of the target to be processed; The determining whether to remove the trajectory information of the target to be processed from the target tracking list based on the trajectory information of the target to be processed and the latest loss threshold includes: Based on the magnitude relationship between the loss count of the target to be processed and the latest loss threshold, determine whether to remove the trajectory information of the target to be processed from the target tracking list; In response to the loss count of the target to be processed being not less than the latest loss threshold, remove the trajectory information of the target to be processed from the target tracking list.

11. An electronic device, characterized in that, Comprising a memory and a processor coupled to each other, wherein the memory stores program instructions, and the processor is configured to execute the program instructions to implement the target tracking method according to any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that, Stored with program instructions that can be run by a processor, the program instructions are used to implement the target tracking method according to any one of claims 1 to 10.

Images