A tracking method, device, and computer storage medium for a target object

Abstract

The present application relates to the technical field of vehicle control, and particularly relates to a method, device and computer storage medium for tracking a target object; the tracking method includes: obtaining a current image frame group in an image frame sequence; respectively comparing multiple non-initial image frames with an initial image frame to obtain motion direction change regions corresponding to the multiple non-initial image frames; updating information of the terminal image frame in the current image frame group based on the regional distribution of the motion direction change regions to obtain an updated terminal image frame; performing target tracking on the target object based on the non-terminal image frames in the current image frame group and the updated terminal image frame to obtain a tracking result of the target object; by comparing information of multiple non-initial image frames with the initial image frame and updating information of the terminal image frame based on the regional distribution result of the comparison, the jitter of the target object information in the current image frame group is reduced, and the tracking accuracy of the target object is improved.

Description

Technical Field

[0001] This application relates to the technical field of vehicle control, and particularly relates to a method, device, and computer storage medium for tracking a target object. Background Art

[0002] When a vehicle is in an intelligent driving state, it is necessary for the vehicle to track a target object to avoid collisions between the vehicle and the target object during intelligent driving. However, during the process of tracking the target object, due to the occlusion of other target objects or changes in the acquisition angle, the orientation of the target object jitters, resulting in inaccurate tracking of the target object. Summary of the Invention

[0003] In view of the above problems in the prior art, the purpose of this application is to compare the information of multiple non-initial image frames with that of the initial image frame, and update the information of the end image frame based on the comparison result of the regional distribution, so as to reduce the jitter of the target object information in the current image frame group, and further improve the tracking accuracy of the target object.

[0004] To solve the above problems, this application provides a method for tracking a target object, including:

[0005] Obtain the current image frame group in the image frame sequence; the current image frame group includes an initial image frame and multiple non-initial image frames; each image frame in the image frame sequence includes target object information of the target object;

[0006] Compare the information of the multiple non-initial image frames with that of the initial image frame respectively to obtain the motion direction change regions corresponding to the multiple non-initial image frames;

[0007] Update the information of the end image frame in the current image frame group based on the regional distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain an updated end image frame;

[0008] Perform target tracking on the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the tracking result of the target object.

[0009] In an embodiment of this application, the target object has motion orientation information. The step of comparing the information of the multiple non-initial image frames with that of the initial image frame respectively to obtain the motion direction change regions corresponding to the multiple non-initial image frames includes:

[0010] Perform an orientation angle change operation on the motion orientation information corresponding to the multiple non-initial image frames and the motion orientation information corresponding to the initial image frame to obtain the orientation differences corresponding to the multiple non-initial image frames;

[0011] Determine the motion direction change regions corresponding to the multiple non-initial image frames based on the correspondence between the motion change regions and the orientation differences, and the orientation differences corresponding to the multiple non-initial image frames.

[0012] In the embodiments of the present application, the number of motion direction change regions corresponding to the multiple non-initial image frames is multiple; the region distribution includes the number of image frames corresponding to the multiple motion direction change regions; the information update of the end image frame in the current image frame group based on the region distribution of the motion direction change regions corresponding to the multiple non-initial image frames respectively includes:

[0013] Based on the number of image frames corresponding to the multiple motion direction change regions, determine a target region from the multiple motion direction change regions; the number of image frames corresponding to the target region is greater than the number of image frames corresponding to other regions; the other regions are the regions other than the target region among the multiple motion direction change regions;

[0014] Determine a target image frame from the non-initial image frames corresponding to the target region; the target image frame is a frame in the region image frames that is close to the end image frame;

[0015] Update the motion orientation information corresponding to the end image frame based on the motion orientation information corresponding to the target image frame to obtain the updated end image frame.

[0016] In the embodiments of the present application, the determining of the target image frame from the non-initial image frames corresponding to the target region includes:

[0017] In the case where the multiple non-initial image frames respectively correspond to different motion direction change regions, determine the adjacent image frame as the target image frame; the adjacent image frame is the previous image frame of the end image frame.

[0018] In the embodiments of the present application, the target tracking of the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the tracking result of the target object includes:

[0019] Update the target object information of the target object by Kalman filtering based on the non-end image frames in the current image frame group and the updated end image frame to obtain the updated target object information corresponding to the current image frame group;

[0020] Perform target tracking on the target object based on the updated target object information corresponding to the current image frame group to obtain the tracking result of the target object.

[0021] In the embodiments of the present application, before obtaining the current image frame group in the image frame sequence, the method includes:

[0022] Obtain the captured image frames;

[0023] Pre-screen the captured image frames to obtain screened image frames; the screened image frames include the target object information;

[0024] Determine the image frame sequence based on the screened image frames.

[0025] In the embodiments of the present application, the determining the image frame sequence based on the screened image frames includes:

[0026] Perform Kalman filtering on the screened image frames based on a preset noise parameter and a preset covariance parameter to obtain initialized image frames;

[0027] Generate the image frame sequence based on the initialized image frames.

[0028] On the other hand, the present application also provides a tracking device for a target object, and the device includes:

[0029] An acquisition module, configured to acquire a current image frame group in the image frame sequence; the current image frame group includes an initial image frame and multiple non-initial image frames; the image frames in the image frame sequence all include the target object information of the target object;

[0030] An information comparison module, configured to respectively compare the multiple non-initial image frames with the initial image frame to obtain the motion direction change regions corresponding to the multiple non-initial image frames;

[0031] An information update module, configured to update the information of the end image frame in the current image frame group based on the region distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain an updated end image frame;

[0032] A target tracking module, configured to perform target tracking on the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the tracking result of the target object.

[0033] On the other hand, the present application also provides an electronic device, and the device includes a processor and a memory, and at least one instruction or at least one program segment is stored in the memory, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement the above-mentioned tracking method for a target object.

[0034] On the other hand, the present application also provides a computer storage medium, in which at least one instruction or at least one program segment is stored, and the at least one instruction or the at least one program segment is loaded and executed by a processor to implement the tracking method of the target object as described above.

[0035] Due to the above technical solution, the tracking method of the target object described in the present application has the following beneficial effects:

[0036] By respectively comparing the information of multiple non-initial image frames in the current image frame group with the initial image frame, the motion change regions corresponding to the multiple non-initial image frames are obtained; and based on the regional distribution of the motion change regions corresponding to the multiple non-initial image frames, the information of the end image frame is updated, where the end image frame is the image frame corresponding to the current moment; thereby making the difference in the target object information corresponding to the end image frame and the target object information in the current image frame group small, thereby reducing the jitter of the target object information in the current image frame group, and thereby improving the tracking accuracy of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solution of the present application, the accompanying drawings required for the implementation examples or the description of the prior art will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0038] Figure 1 is a schematic flowchart of a method for tracking a target object provided by an embodiment of the present application;

[0039] Figure 2 is a schematic flowchart for determining a motion direction change region in a method for tracking a target object provided by an embodiment of the present application;

[0040] Figure 3 is a schematic flowchart for updating an end image frame in a method for tracking a target object provided by an embodiment of the present application;

[0041] Figure 4 is a schematic flowchart for determining a tracking result in a method for tracking a target object provided by an embodiment of the present application;

[0042] Figure 5 is a schematic flowchart for determining an image frame sequence in a method for tracking a target object provided by an embodiment of the present application;

[0043] Figure 6 is a schematic flowchart for determining an image frame sequence in a method for tracking a target object provided by an embodiment of the present application;

[0044] Figure 7 It is a schematic structural diagram of a tracking device for a target object provided by an embodiment of the present application;

[0045] Figure 8 It is a hardware structure block diagram of a tracking method for a target object provided by an embodiment of the present application. Detailed implementation manners

[0046] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0047] As used herein, the term "one embodiment" or "embodiment" refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present application. In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "top", "bottom", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present application. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. Moreover, the terms "first", "second", etc. are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order different from those illustrated or described here.

[0048] Combined with Figure 1 , a tracking method for a target object provided by an embodiment of the present application is introduced. The method includes:

[0049] S1001. Obtain the current image frame group in the image frame sequence; the current image frame group includes an initial image frame and multiple non-initial image frames; all the image frames in the image frame sequence include the target object information of the target object; the image frame sequence refers to a sequence that continuously adds image frames based on the passage of time; the current image frame group refers to a group of image frames corresponding to the current time node. Among them, in the current image frame group, the image frame with the earliest time node is the initial image frame, and the remaining image frames except the initial image frame are non-initial image frames; that all the image frames in the image frame sequence include the target object information of the target object indicates that the mode of tracking the target object has been entered; the target object includes but is not limited to a moving vehicle, a pedestrian, and a roadblock; the target object information includes but is not limited to the movement orientation information of the target, the movement speed of the target object, and the shape of the target object.

[0050] In a specific embodiment of the present application, the current image frame group may be 6 image frames or 8 image frames, and the specific selection is based on actual needs and computing needs.

[0051] S1002. Compare the multiple non-initial image frames with the initial image frame respectively to obtain the motion direction change regions corresponding to the multiple non-initial image frames; the information comparison means comparing the target object information corresponding to the non-initial image with the target object information corresponding to the initial image frame; the motion direction change region characterizes the change amplitude of the movement orientation of the target object.

[0052] S1003. Update the information of the end image frame in the current image frame group based on the regional distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain the updated end image frame; the regional distribution of the motion direction change regions refers to the distribution of non-initial image frames in different motion direction change regions; in the current image frame group, the image frame with the latest time node is the end image frame, and the remaining image frames except the end image frame are non-end image frames; the information update means correcting the target object information corresponding to the image frame. Specifically, it can be information replacement or information fusion.

[0053] S1004. Perform target tracking on the target object based on the non-end image frames and the updated end image frame in the current image frame group to obtain the tracking result of the target object; the target tracking can characterize predicting the motion trajectory of the target object, and the tracking result can be the motion trajectory of the target object or the motion conflict relationship between the current object and the target object. Here, the current object is the object running this method, and the current object includes but is not limited to a vehicle and a drone. The motion conflict relationship characterizes whether there is a possibility of collision or encounter between the current object and the target object.

[0054] In the embodiments of the present application, by respectively comparing the information of multiple non-initial image frames in the current image frame group with the initial image frame, the motion change regions corresponding to each of the multiple non-initial image frames are obtained; and based on the regional distribution of the motion change regions corresponding to each of the multiple non-initial image frames, the information of the end image frame is updated, where the end image frame is the image frame corresponding to the current moment; thereby making the difference in the target object information corresponding to the end image frame and the target object information in the current image frame group small, thereby reducing the jitter of the target object information in the current image frame group, and thereby improving the tracking accuracy of the target object.

[0055] Reference Figure 2 , in the embodiments of the present application, the target object has motion orientation information, and the motion orientation information represents the orientation of the target object. Taking a vehicle as an example, the motion orientation information is the direction pointed by the vehicle head; S1002 includes:

[0056] S2001. Perform an orientation angle change operation on the motion orientation information corresponding to each of the multiple non-initial image frames and the motion orientation information corresponding to the initial image frame to obtain the orientation differences corresponding to the multiple non-initial image frames; the orientation angle change operation refers to calculating the orientation change angle of the target object; the orientation difference represents the orientation change of the target object.

[0057] In the specific embodiments of the present application, the following formula can be used for the orientation angle change operation:

[0058] ID = (heading[i] – heading[0]) / (2 × π / 36)

[0059] where ID represents the orientation difference; heading[i] represents the motion orientation information corresponding to any non-initial image frame, and the motion orientation information can be angle information; heading[0] represents the orientation value corresponding to the initial image frame.

[0060] S2002. Based on the correspondence between the motion change region and the orientation difference and the orientation differences corresponding to the multiple non-initial image frames, determine the motion direction change regions corresponding to each of the multiple non-initial image frames; specifically, any motion change region corresponds to at least one orientation difference value range; the orientation difference value range includes a continuous plurality of orientation differences.

[0061] In the embodiments of the present application, the motion change region is N regions (N ≥ 2), and the entire value range of the orientation value is A0 - A N ; then, the Nth region corresponds to A N -A N-1 and A0 - A1; the (N - 1)th region corresponds to A N-1 -A N-2and A1 - A2; and so on until N regions divide A0 - A N are completely divided.

[0062] In a specific embodiment of the present application, the motion change region is divided into six regions, i.e., N = 6; the entire value range of the orientation difference is from 0 to 35, i.e., A0 = 0, A N = 35; then, the first region corresponds to the orientation difference range of 33 - 35 and the orientation difference range of 0 - 2; the second region corresponds to the orientation difference range of 30 - 32 and the orientation difference range of 3 - 5; the third region corresponds to the orientation difference range of 27 - 29 and the orientation difference range of 6 - 8; the fourth region corresponds to the orientation difference range of 24 - 26 and the orientation difference range of 9 - 11; the fifth region corresponds to the orientation difference range of 21 - 23 and the orientation difference range of 12 - 14; the sixth region corresponds to the orientation difference range of 15 - 20.

[0063] In other embodiments of the present application, the number of motion change regions varies based on actual requirements, and the entire value range of the orientation difference varies with the change of the formula.

[0064] In the embodiments of the present application, by using the orientation angle change operation and the correspondence between the motion change region and the orientation difference, the motion change region corresponding to each non - initial image frame is determined, which improves the stability and reliability of determining the motion direction change region, and further improves the reliability of target tracking.

[0065] Referring to Figure 3 , in the embodiments of the present application, the number of motion direction change regions corresponding to multiple non - initial image frames is multiple; the region distribution includes the number of image frames corresponding to multiple motion direction change regions; S1003 includes:

[0066] S3001. Based on the number of image frames corresponding to multiple motion direction change regions, determine the target region from multiple motion direction change regions; the number of image frames corresponding to the target region is greater than the number of image frames corresponding to other regions; other regions are regions other than the target region among multiple motion direction change regions.

[0067] In the embodiments of the present application, the target motion change region may include one type of motion change region or multiple types of motion change regions; specifically, assuming that the current image frame group is M frames, excluding the initial image frame, there are M - 1 frames in the N motion change regions. If the number of image frames in the first motion change region among the N motion direction change regions is X frames, and the number of image frames in the remaining motion change regions is less than X, then the first motion change region is the target motion change region; if the number of image frames in the first motion change region and the third motion change region among the N motion direction change regions is both Y frames, then both the first motion change region and the third motion change region are target motion change regions.

[0068] In a specific embodiment of the present application, when N = 6, M can be 6, then X is 3, and Y is 2.

[0069] S3002. Determine a target image frame from the non-initial image frames corresponding to the target region; the target image frame is a frame close to the end image frame among the region image frames.

[0070] S3003. Update the motion orientation information corresponding to the end image frame based on the motion orientation information corresponding to the target image frame to obtain an updated end image frame.

[0071] In the embodiments of the present application, by determining the target motion change region through voting based on the number of image frames corresponding to the motion direction change region, it is possible to ensure that the motion change region in the current image frame group is in a stable state, thereby making the difference in the target object information corresponding to the end image frame and the target object information in the current image frame group small, reducing the jitter of the target object information in the current image frame group, and improving the tracking accuracy of the target object.

[0072] In the embodiments of the present application, S3002 includes:

[0073] In the case where each of the multiple non-initial image frames corresponds to a different motion direction change region, determine the adjacent image frame as the target image frame; the adjacent image frame is the previous image frame of the end image frame; each of the multiple non-initial image frames corresponding to a different motion direction change region indicates that the orientation change of the target object is variable.

[0074] In the embodiments of the present application, by determining the adjacent image frame as the target image frame, the orientation jitter of the target object is reduced, and the tracking accuracy of the target object is improved.

[0075] Reference Figure 4 In the embodiments of the present application, S1004 includes:

[0076] S4001. Update the target object information of the target object by performing Kalman filtering on the non-terminal image frames in the current image frame group and the updated terminal image frame, so as to obtain the updated target object information corresponding to the current image frame group.

[0077] S4002. Perform target tracking on the target object based on the updated target object information corresponding to the current image frame group to obtain the tracking result of the target object.

[0078] In the embodiment of the present application, by using Kalman filtering for update prediction, the tracking accuracy of the target object can be improved.

[0079] Reference Figure 5 , before S1001 in the embodiment of the present application, the target object tracking method further includes:

[0080] S5001. Obtain the acquired image frame; the acquired image frame is the image frame acquired based on the image acquisition device.

[0081] S5002. Perform pre-screening on the acquired image frame to obtain the screened image frame; the screened image frame includes the target object information.

[0082] S5003. Determine the image frame sequence based on the screened image frame.

[0083] In the embodiment of the present application, through pre-screening, the image frames corresponding to the target object information are screened to obtain the screened image frames, and the image frame sequence is generated based on the screened image frames, so as to ensure continuous tracking of the target object, and further improve the tracking reliability of the target object.

[0084] In a specific embodiment of the present application, the image frame sequence includes at least 3 screened image frames to ensure continuous acquisition of the target object, and further avoid tracking of invalid target objects.

[0085] In the embodiment of the present application, within a preset time period, when the number of image frames in the image frame sequence corresponding to the target object is less than the preset number, stop tracking the target object; by limiting the number of target object image frames in the image frame sequence, further avoid tracking the target object when the target object is in an occluded and lost state.

[0086] Reference Figure 6 , in the embodiment of the present application, S5003 includes:

[0087] S6001. Perform Kalman filtering on the screened image frame based on the preset noise parameter and the preset covariance parameter to obtain the initialized image frame; the target object information in the initialized image frame is the processing data information in the subsequent tracking process.

[0088] S6002. Generate an image frame sequence based on an initial image frame.

[0089] In the embodiments of the present application, the image frame is initialized by using Kalman filtering, thereby improving the information accuracy of the target object information and further improving the tracking accuracy of the target tracking.

[0090] Reference Figure 7 , the embodiments of the present application further provide a tracking device for a target object, and the device includes:

[0091] An acquisition module 101, configured to acquire a current image frame group in the image frame sequence; the current image frame group includes an initial image frame and multiple non-initial image frames; the image frames in the image frame sequence all include target object information of the target object;

[0092] An information comparison module 102, configured to respectively compare multiple non-initial image frames with the initial image frame to obtain motion direction change regions corresponding to the multiple non-initial image frames;

[0093] An information update module 103, configured to update the information of the end image frame in the current image frame group based on the region distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain an updated end image frame;

[0094] A target tracking module 104, configured to perform target tracking on the target object based on the non-end image frames and the updated end image frame in the current image frame group to obtain a tracking result of the target object.

[0095] The target object has motion orientation information, and the information comparison module includes:

[0096] A change operation unit, configured to perform an orientation angle change operation on the motion orientation information corresponding to the multiple non-initial image frames and the motion orientation information corresponding to the initial image frame to obtain orientation differences corresponding to the multiple non-initial image frames;

[0097] A change determination unit, configured to determine the motion direction change regions corresponding to the multiple non-initial image frames based on the correspondence between the motion change regions and the orientation differences and the orientation differences corresponding to the multiple non-initial image frames.

[0098] The number of motion direction change regions corresponding to the multiple non-initial image frames is multiple; the region distribution includes the number of image frames corresponding to the multiple motion direction change regions; the information update module includes:

[0099] A target screening unit determines a target area from multiple motion direction change areas based on the number of image frames corresponding to the multiple motion direction change areas; the number of image frames corresponding to the target area is greater than the number of image frames corresponding to other areas; the other areas are the areas other than the target area among the multiple motion direction change areas;

[0100] A target determination unit is used to determine a target image frame from the non-initial image frames corresponding to the target area; the target image frame is a frame of the area image frames that is close to the end image frame;

[0101] An update determination unit is used to update the motion orientation information corresponding to the end image frame based on the motion orientation information corresponding to the target image frame to obtain an updated end image frame.

[0102] The target determination unit includes:

[0103] A target determination subunit is used to determine adjacent image frames as target image frames when each of the multiple non-initial image frames corresponds to a different motion direction change area; the adjacent image frames are the previous image frame of the end image frame.

[0104] The target tracking module includes:

[0105] A first Kalman filtering unit is used to perform Kalman filtering update on the target object information of the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the updated target object information corresponding to the current image frame group;

[0106] A target tracking unit is used to perform target tracking on the target object based on the updated target object information corresponding to the current image frame group to obtain the tracking result of the target object.

[0107] The tracking device of the target object further includes:

[0108] An acquisition and obtaining module is used to obtain acquisition image frames;

[0109] A pre-screening module is used to perform pre-screening on the acquisition image frames to obtain screened image frames; the screened image frames include target object information;

[0110] A sequence generation module is used to determine an image frame sequence based on the screened image frames.

[0111] The sequence generation module includes:

[0112] A second Kalman filtering unit is used to perform Kalman filtering on the screened image frames based on preset noise parameters and preset covariance parameters to obtain initialized image frames;

[0113] A sequence generation unit is used to generate an image frame sequence based on the initialized image frames.

[0114] An embodiment of this application also provides an electronic device, which includes a processor and a memory. At least one instruction or at least one program segment is stored in the memory, and the at least one instruction or the at least one program segment is loaded and executed by the processor to implement the tracking method of the target object as described above.

[0115] The memory can be used to store software programs and modules. The processor runs the software programs and modules stored in the memory to execute various functional applications and data processing. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include high-speed random access memory and can also include non-volatile memory, such as at least one hard disk storage device, a flash memory device, or other volatile solid-state storage devices. Correspondingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0116] The method embodiment provided by the embodiment of this application can be executed in an electronic device such as a mobile terminal, a computer terminal, a server, or a similar computing device. Figure 8 This is the electronic device provided by the embodiment of this application. As Figure 8 shown, the electronic device 900 can have relatively large differences due to different configurations or performances, and can include one or more central processing units (CPUs) 910 (the processor 910 can include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 930 for storing data, and one or more storage media 920 (such as one or more mass storage devices) for storing application programs 923 or data 922. Among them, the memory 930 and the storage media 920 can be transient storage or persistent storage. The program stored in the storage media 920 can include one or more modules, and each module can include a series of instruction operations on the electronic device. Further, the central processor 910 can be set to communicate with the storage media 920 and execute a series of instruction operations in the storage media 920 on the electronic device 900. The electronic device 900 can also include one or more power supplies 960, one or more wired or wireless network interfaces 950, one or more input / output interfaces 940, and / or one or more operating systems 921, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0117] The input / output interface 940 can be used to receive or transmit data via a network. Specific examples of the aforementioned network may include a wireless network provided by a communication provider of the electronic device 900. In one example, the input / output interface 940 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the input / output interface 940 can be a Radio Frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0118] Those of ordinary skill in the art can understand that Figure 8 the structure shown is only illustrative and does not limit the structure of the aforementioned electronic device. For example, the electronic device 900 may further include more or fewer components than Figure 8 those shown, or have a different configuration from Figure 8 those shown.

[0119] An embodiment of the present application also provides a storage medium, in which at least one instruction or at least one segment of program is stored, and the at least one instruction or at least one segment of program is loaded and executed by a processor to implement the tracking method of the target object as described above.

[0120] The above description has fully disclosed the specific implementation manners of the present application. It should be noted that any changes made by those skilled in the art to the specific implementation manners of the present application do not depart from the scope of the claims of the present application. Accordingly, the scope of the claims of the present application is not limited solely to the foregoing specific implementation manners.

Claims

1. A tracking method for a target object, characterized in that Including: Obtain the current image frame group in the image frame sequence; The current image frame group includes an initial image frame and multiple non-initial image frames; Each image frame in the image frame sequence includes target object information of the target object; Compare the information of the multiple non-initial image frames with the information of the initial image frame respectively to obtain the motion direction change regions corresponding to the multiple non-initial image frames; Update the information of the end image frame in the current image frame group based on the region distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain the updated end image frame; Perform target tracking on the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the tracking result of the target object; The target object has motion orientation information, and the number of motion direction change regions corresponding to the multiple non-initial image frames is multiple; the region distribution includes the number of image frames corresponding to the multiple motion direction change regions; The information update includes: Based on the number of image frames corresponding to the multiple motion direction change regions, determine a target region from the multiple motion direction change regions; the number of image frames corresponding to the target region is greater than the number of image frames corresponding to other regions; the other regions are the regions other than the target region among the multiple motion direction change regions; Determine a target image frame from the non-initial image frames corresponding to the target region; the target image frame is a frame in the region image frames that is close to the end image frame; Update the motion orientation information corresponding to the end image frame based on the motion orientation information corresponding to the target image frame to obtain the updated end image frame.

2. The tracking method of a target object according to claim 1, wherein The step of comparing the information of the multiple non-initial image frames with the information of the initial image frame respectively to obtain the motion direction change regions corresponding to the multiple non-initial image frames includes: Perform an orientation angle change operation on the motion orientation information corresponding to the multiple non-initial image frames and the motion orientation information corresponding to the initial image frame to obtain the orientation differences corresponding to the multiple non-initial image frames; Determine the motion direction change regions corresponding to the multiple non-initial image frames based on the correspondence between the motion change regions and the orientation differences and the orientation differences corresponding to the multiple non-initial image frames.

3. The tracking method of a target object according to claim 1, characterized in that, The step of determining a target image frame from the non-initial image frames corresponding to the target region includes: When the multiple non-initial image frames respectively correspond to different motion direction change regions, determine the adjacent image frame as the target image frame; the adjacent image frame is the previous image frame of the end image frame.

4. The tracking method of a target object according to claim 1, characterized in that, The step of performing target tracking on the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the tracking result of the target object includes: Perform Kalman filter update on the target object information of the target object based on the non-end image frames in the current image frame group and the updated end image frame to obtain the updated target object information corresponding to the current image frame group; Perform target tracking on the target object based on the updated target object information corresponding to the current image frame group to obtain the tracking result of the target object.

5. The tracking method of a target object according to claim 1, characterized in that, Before obtaining the current image frame group in the image frame sequence, the method includes: Obtain the captured image frames. Perform pre-screening on the captured image frames to obtain the screened image frames; the screened image frames include the target object information. Determine the image frame sequence based on the screened image frames.

6. The tracking method of a target object according to claim 5, wherein The determining the image frame sequence based on the screened image frames includes: Perform Kalman filtering on the screened image frames based on a preset noise parameter and a preset covariance parameter to obtain the initialized image frames. Generate the image frame sequence based on the initialized image frames.

7. A tracking device for a target object, characterized in that, Includes: An acquisition module, configured to acquire the current image frame group in the image frame sequence. The current image frame group includes an initial image frame and multiple non-initial image frames. The image frames in the image frame sequence all include the target object information of the target object. An information comparison module, configured to respectively compare the multiple non-initial image frames with the initial image frame to obtain the motion direction change regions corresponding to the multiple non-initial image frames. An information update module, configured to update the information of the last image frame in the current image frame group based on the regional distribution of the motion direction change regions corresponding to the multiple non-initial image frames to obtain the updated last image frame. A target tracking module, configured to perform target tracking on the target object based on the non-last image frames and the updated last image frame in the current image frame group to obtain the tracking result of the target object. The target object has motion orientation information, and the number of motion direction change regions corresponding to the multiple non-initial image frames is multiple; the regional distribution includes the number of image frames corresponding to the multiple motion direction change regions. The information update includes: Based on the number of image frames corresponding to the multiple motion direction change regions, determine a target region from the multiple motion direction change regions; the number of image frames corresponding to the target region is greater than the number of image frames corresponding to other regions; the other regions are the regions other than the target region among the multiple motion direction change regions. Determine a target image frame from the non-initial image frames corresponding to the target region; the target image frame is a frame in the regional image frames that is close to the last image frame. Update the motion orientation information corresponding to the last image frame based on the motion orientation information corresponding to the target image frame to obtain the updated last image frame.

8. A computer storage medium, characterized in that, At least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by a processor to implement the tracking method of the target object as described in any one of claims 1-6.

9. An electronic device, characterized in that, The device includes a processor and a memory, and at least one instruction or at least one program is stored in the memory, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the tracking method of the target object as described in any one of claims 1-6.

Images