Target tracking method and related device thereof

Abstract

The invention relates to a target tracking method and a related device thereof, and the method comprises the steps: obtaining a first image collected by a visual sensor in the driving process of a vehicle, and the first image comprises a target obstacle; and when the target obstacle is in a shielded state, predicting a motion track of the target obstacle according to the first image and a particle filtering algorithm. According to the scheme provided by the invention, the movement track of the blocked target in the blocked process can be predicted, the collision risk of the vehicle and the obstacle is reduced, and the safety is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a target tracking method and related devices. Background Art

[0002] Target tracking systems play a big role in many fields, such as security, traffic control, etc. However, due to the rapid changes and complexity of pedestrian movements, multi-person tracking is very difficult.

[0003] When a car is driving, it is easy to encounter pedestrians that are partially obscured, causing the driver to lose target tracking. The current target tracking method cannot predict the target's movement trajectory well, and cannot assist the driver to know the approximate movement trajectory of the obscured target, which is prone to collision risks. Summary of the invention

[0004] In order to solve or partially solve the problems existing in the related technology, the present application provides a target tracking method and related devices, which can predict the movement trajectory of a blocked target during the blocking process, reduce the risk of collision between the vehicle and the obstacle, and improve safety.

[0005] A first aspect of the present application provides a target tracking method, comprising acquiring a first image captured by a visual sensor of a vehicle during driving, the first image including a target obstacle; when the target obstacle is in an obscured state, predicting a motion trajectory of the target obstacle based on the first image and a particle filtering algorithm.

[0006] In combination with the first aspect, in a possible implementation manner of the first aspect, predicting the motion trajectory of the target obstacle based on the first image and the particle filtering algorithm includes: acquiring first state parameters and a preset tracking period of the target obstacle in the first image, the first state parameters including a first position and a first speed; inputting the first state parameters and the preset tracking period into a preset state transfer equation to obtain second state parameters of the target obstacle, the second state parameters including a second position and a second speed; and generating the motion trajectory of the target obstacle based on the first state parameters and the second state parameters.

[0007] In combination with the first aspect, in a possible implementation manner of the first aspect, the formula of the state transfer equation is as follows:

[0008]

[0009] in, xi ,y i are the horizontal and vertical coordinates of the second position, respectively, v x 、v yare the lateral speed of the second speed and the longitudinal speed of the second speed respectively, Δt is the preset tracking period, w x 、w y , is the process noise, x i-1 ,y i-1 are the abscissa and ordinate of the first position, is the lateral speed of the first speed and the longitudinal speed of the first speed.

[0010] In combination with the first aspect, in a possible implementation method of the first aspect, it also includes: acquiring a second image after the first image, and calculating the displacement between the first image and the second image; if the displacement is less than a second threshold, determining that the target obstacle is in an obstructed state; if the displacement is greater than or equal to the second threshold, determining that the target obstacle is not in an obstructed state.

[0011] In combination with the first aspect, in a possible implementation of the first aspect, the calculating the displacement between the first image and the second image includes: extracting feature points of the target obstacle from the first image and the second image respectively; and calculating the displacement between the two images based on the feature points of at least two images.

[0012] In combination with the first aspect, in a possible implementation manner of the first aspect, extracting feature points of the target obstacle from the first image and the second image respectively includes: for any one of the first image and the second image, extracting brightness data of all pixels of the target obstacle from the image, and taking pixels whose brightness data difference with adjacent pixels is greater than a first threshold as target pixels; and comparing the target pixels with feature pixels of the target obstacle pre-stored in a database, so as to take target pixels matching the feature pixels as the feature points of the target obstacle.

[0013] In combination with the first aspect, in a possible implementation manner of the first aspect, it also includes: obtaining a third image of the target obstacle after being blocked; obtaining feature points of the target obstacle in the third image; comparing the number of feature points in the third image with the number of feature points in the first image; if the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold, determining that the target obstacle in the third image is the same as the target obstacle in the first image.

[0014] A second aspect of the present application provides a target tracking device, including an acquisition module, which acquires a first image collected by a visual sensor of a vehicle during driving, wherein the first image includes a target obstacle;

[0015] A prediction module predicts a motion trajectory of the target obstacle according to the first image and a particle filter algorithm when the target obstacle is in an obstructed state.

[0016] In combination with the second aspect, in a possible implementation manner of the second aspect, the prediction module is further used to obtain a first state parameter and a preset tracking period of the target obstacle in the first image, the first state parameter including a first position and a first speed; input the first state parameter and the preset tracking period into a preset state transfer equation to obtain a second state parameter of the target obstacle, the second state parameter including a second position and a second speed; and generate a motion trajectory of the target obstacle according to the first state parameter and the second state parameter.

[0017] In combination with the second aspect, in a possible implementation method of the second aspect, the acquisition module is also used to acquire a second image after the first image, and calculate the displacement between the first image and the second image; if the displacement is less than a second threshold, it is determined that the target obstacle is in an obscured state; if the displacement is greater than or equal to the second threshold, it is determined that the target obstacle is not in an obscured state.

[0018] In combination with the second aspect, in a possible implementation of the second aspect, the acquisition module is also used to extract feature points of the target obstacle from the first image and the second image respectively; and calculate the displacement between the two images based on the feature points of at least two images.

[0019] In combination with the second aspect, in a possible implementation manner of the second aspect, the acquisition module is further used to extract brightness data of all pixels of the target obstacle from any one of the first image and the second image, and take pixels whose brightness data difference with adjacent pixels is greater than a first threshold as target pixels; compare the target pixels with feature pixels of the target obstacle pre-stored in a database to take target pixels that match the feature pixels as feature points of the target obstacle.

[0020] In combination with the second aspect, in a possible implementation manner of the second aspect, the acquisition module is further used to acquire a third image of the target obstacle after being blocked; acquire feature points of the target obstacle in the third image; compare the number of feature points in the third image with the number of feature points in the first image; if the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold, determine that the target obstacle in the third image is the same as the target obstacle in the first image.

[0021] A third aspect of the present application provides an electronic device, including:

[0022] Processor; and

[0023] The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method as described above.

[0024] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0025] A fifth aspect of the present application provides a computer program product, comprising a computer program / instruction, which implements the method described above when executed by a processor.

[0026] The technical solution provided by this application may have the following beneficial effects:

[0027] The target tracking method and related devices of the present application include obtaining a first image collected by a visual sensor of a vehicle during driving, wherein the first image includes a target obstacle; when the target obstacle is in an obstructed state, the motion trajectory of the target obstacle is predicted based on the first image and a particle filtering algorithm, and the motion trajectory of the obstructed target during the obstruction process can be predicted, thereby reducing the risk of collision between the vehicle and the obstacle and improving safety.

[0028] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0030] Figure 1 is a flowchart of a target tracking method shown in an embodiment of the present application;

[0031] Figure 2 is a schematic diagram of the structure of a target tracking device shown in an embodiment of the present application;

[0032] Figure 3 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0033] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0034] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "a", "an" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in this article refers to and includes any or all possible combinations of one or more associated listed items.

[0035] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0036] In related technologies, when vehicles travel in complex and ever-changing traffic environments, some obstacle targets are often blocked by other vehicles, buildings or vegetation, which not only causes the target to temporarily disappear in the visual sensor, but also increases the complexity of the tracking algorithm. Occlusion events require the tracking system to have a highly sensitive detection mechanism to quickly identify the occurrence of occlusion and predict the possible trajectory of the target after occlusion. At the same time, the system also needs to keep a memory of pedestrian features during occlusion so that it can quickly recapture the target and restore a stable tracking state after the occlusion ends.

[0037] After the target is blocked, the vehicle and the driver cannot judge the position and speed of the target in real time, which may easily lead to the driver's misjudgment of the vehicle operation and collision with the target. Therefore, by predicting the position and speed of the target after being blocked and predicting the target's motion trajectory, the entire track of the target during the blocking process can be obtained, which is helpful for the driver to judge the next operation.

[0038] In response to the above problems, an embodiment of the present application provides a target tracking method, which can predict the movement trajectory of a blocked target during the blocking process, reduce the risk of collision between the vehicle and the obstacle, and improve safety.

[0039] The technical solution of the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0040] Figure 1 It is a flowchart of the target tracking method shown in an embodiment of the present application.

[0041] See also Figure 1 , a target tracking method, comprising:

[0042] S110: Acquire a first image captured by a visual sensor during the driving of the vehicle, where the first image includes a target obstacle.

[0043] Specifically, during the operation, the driver powers on the vehicle, and the visual sensor installed on the vehicle immediately enters the working state. For example, it can be a visual sensor installed at the lower center edge of the front windshield of the vehicle. If the vehicle status, sensor status and CAN communication (local area network) are normal, the vehicle controller starts to operate. If there is a fault, a fault prompt is given. Multiple visual sensors can be set in the vehicle. When the vehicle is driving, for example, the vehicle is in normal forward driving, the visual sensor obtains the image in front of the vehicle in real time. The first image can be obtained through the visual sensor, and the first image includes the target obstacle. A controller is set in the vehicle, and the first image obtained by the visual sensor can be transmitted to the controller for calculation.

[0044] This embodiment does not limit the type of visual sensor. For example, the visual sensor may be a camera device, such as a network camera, etc. Those skilled in the art should understand that a network camera refers to a new generation of cameras that are produced by combining traditional cameras with network technology. In addition to having the image capture function of a general traditional camera, the camera also has a built-in digital compression controller and a network-based operating system, so that the video data is compressed and encrypted and transmitted to other devices (such as terminals or servers) through a network (such as a local area network or a wireless network, etc.).

[0045] S120: When the target obstacle is in an obstructed state, predicting a movement trajectory of the target obstacle according to the first image and a particle filtering algorithm.

[0046] Specifically, after the first image is sent to the controller, the controller can judge the target obstacle of the first image. When it is judged that the target obstacle is in an obscured state, the motion trajectory of the target obstacle can be predicted based on the first image and the particle filtering algorithm. The motion trajectory of the target obstacle may include parameters such as the position and speed of the target obstacle in multiple time frames.

[0047] In a possible implementation, first state parameters and a preset tracking period of a target obstacle in a first image are obtained, the first state parameters including a first position and a first speed; the first state parameters and the preset tracking period are input into a preset state transfer equation to obtain second state parameters of the target obstacle, the second state parameters including a second position and a second speed; and a motion trajectory of the target obstacle is generated according to the first state parameters and the second state parameters.

[0048] Specifically, the target obstacle has a first state parameter in the first image, which can be the first position and first speed of the target obstacle. The preset tracking period can be the acquisition time interval between the first image and the adjacent frame images, which can be preset, such as 0.1s, 0.5s, etc. A group of particles are generated using a random sampling method. Each particle represents a possible state of the target obstacle, and the same initial weight is assigned to each particle, indicating that all particles have the same importance at the initial moment. According to the system motion model, a uniform motion model is used to predict the state of each particle; the first state parameter and the preset tracking period are input into the preset state transfer equation to obtain the second state parameter of the target obstacle, and the motion trajectory of the target obstacle can be generated according to the first state parameter and the second state parameter.

[0049] In a possible implementation, the formula of the state transition equation is as follows:

[0050]

[0051] in, xi ,y i are the horizontal and vertical coordinates of the second position, respectively, v x 、v y are the lateral speed of the second speed and the longitudinal speed of the second speed respectively, Δt is the preset tracking period, w x 、w y , is the process noise, x i-1 ,y i-1 are the abscissa and ordinate of the first position, is the lateral speed of the first speed and the longitudinal speed of the first speed.

[0052] Specifically, the observation model can be used to evaluate the consistency of the first velocity and the first position of each particle with the target obstacle. The weights of all particles can be normalized to ensure that their sum is 1, and resampled according to the weights of the particles. Particles with high weights are more likely to be selected and retained, while particles with low weights are more likely to be deleted. The purpose of resampling is to reduce the number of particles with lower weights and increase the number of particles with higher weights, thereby improving the ability of the particle set to approximate the true state. The state and weight of the particles are continuously updated until a predetermined number of iterations is reached or a certain convergence condition is met. At the end of each time step, the current state of the target obstacle (such as position, speed, etc.) is estimated based on the weight and state information of the particle set.

[0053] In a possible implementation, a second image is acquired after the first image, and a displacement between the first image and the second image is calculated; if the displacement is less than a second threshold, it is determined that the target obstacle is in an obstructed state; if the displacement is greater than or equal to the second threshold, it is determined that the target obstacle is not in an obstructed state.

[0054] Specifically, the second image may be an image after adjacent frames to the first image, and the displacement between the target obstacle in the first image and the target obstacle in the second image may be calculated. If the displacement is less than a second threshold, it may be determined that the target obstacle is in an obstructed state.

[0055] In a possible implementation, feature points of the target obstacle are extracted from the first image and the second image respectively; and the displacement between the two images is calculated based on the feature points of at least two images.

[0056] In a possible implementation, for any one of the first image and the second image, brightness data of all pixels of the target obstacle are extracted from the image, and pixels whose brightness data difference with adjacent pixels is greater than a first threshold are taken as target pixels; the target pixels are compared with feature pixels of the target obstacle pre-stored in a database, so that the target pixels matching the feature pixels are taken as feature points of the target obstacle.

[0057] Specifically, after preprocessing the first image and the second image, the controller can use an algorithm, such as the ORB algorithm, to extract the features of the target obstacle and compare them with the features pre-stored in the database; the first image and the second image can be converted from color images to gray images, and scaled multiple times according to a certain scale factor to generate multiple image layers with different resolutions to handle the target scale change, and the target pixel points are obtained by comparing the brightness difference between all pixels of the target obstacle and its surrounding pixels, and the direction of each target pixel point is calculated by the grayscale centroid method; the image is rotated according to the direction of the target pixel point, and then the grayscale value of the image is compared to generate a binary string descriptor corresponding to the target pixel point; the feature point of the target obstacle is obtained according to the size of the Hamming distance of each target pixel point.

[0058] Specifically, after obtaining the feature points of the target obstacle in the first image, a confidence index can be generated according to the quantity and quality of the feature points. When the confidence index is greater than a preset threshold, it can be preliminarily determined that the target obstacle is in an obscured state. For the first image and the second image, an optical flow algorithm can be used to calculate the displacement of the feature points of the target obstacle. The optical flow results can be filtered or smoothed to reduce the influence of noise and erroneous estimation. When the displacement is greater than the second threshold, it can be determined that the target obstacle is in an obscured state.

[0059] In a possible implementation, a third image of the target obstacle after being blocked is obtained; feature points of the target obstacle in the third image are obtained; the number of feature points in the third image is compared with the number of feature points in the first image; if the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold value, it is determined that the target obstacle in the third image is the same as the target obstacle in the first image.

[0060] Specifically, the visual sensor can obtain a third image of the target obstacle after being blocked, and can obtain feature points of the target obstacle in the third image by the method of obtaining the target obstacle in the first image, and compare the number of feature points in the third image with the number of feature points in the first image; if the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold, it is determined that the target obstacle in the third image is the same as the target obstacle in the first image, and the feature points of the target obstacle in the third image can be used to update the template of the target obstacle in the database.

[0061] The target tracking method and related devices of the present application include obtaining a first image collected by a visual sensor of a vehicle during driving, wherein the first image includes a target obstacle; when the target obstacle is in an obstructed state, the motion trajectory of the target obstacle is predicted based on the first image and a particle filtering algorithm, and the motion trajectory of the obstructed target during the obstruction process can be predicted, thereby reducing the risk of collision between the vehicle and the obstacle and improving safety.

[0062] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides a target tracking device, an electronic device and corresponding embodiments.

[0063] Figure 2 It is a schematic diagram of the structure of the target tracking device 200 shown in an embodiment of the present application.

[0064] See also Figure 2 , a target tracking device 200, comprising:

[0065] The acquisition module 210 acquires a first image collected by a visual sensor during the driving process of the vehicle, where the first image includes a target obstacle.

[0066] Specifically, during the operation, the driver powers on the vehicle, and the visual sensor installed on the vehicle immediately enters the working state. For example, it can be a visual sensor installed at the lower center edge of the front windshield of the vehicle. If the vehicle status, sensor status and CAN communication (local area network) are normal, the vehicle controller starts to operate. If there is a fault, a fault prompt is given. Multiple visual sensors can be set in the vehicle. When the vehicle is driving, for example, the vehicle is in normal forward driving, the visual sensor obtains the image in front of the vehicle in real time. The first image can be obtained through the visual sensor, and the first image includes the target obstacle. A controller is set in the vehicle, and the first image obtained by the visual sensor can be transmitted to the controller for calculation.

[0067] In a possible implementation, the acquisition module 210 is further used to acquire a second image after the first image, and calculate the displacement between the first image and the second image; if the displacement is less than a second threshold, it is determined that the target obstacle is in an obstructed state; if the displacement is greater than or equal to the second threshold, it is determined that the target obstacle is not in an obstructed state.

[0068] In a possible implementation, the acquisition module 210 is further configured to extract feature points of the target obstacle from the first image and the second image respectively; and calculate the displacement between the two images based on the feature points of at least two images.

[0069] In a possible implementation, the acquisition module 210 is further used to extract brightness data of all pixels of the target obstacle from any one of the first image and the second image, and use pixels whose brightness data difference with adjacent pixels is greater than a first threshold as target pixels; and compare the target pixels with feature pixels of the target obstacle pre-stored in a database to use target pixels that match the feature pixels as feature points of the target obstacle.

[0070] In a possible implementation, the acquisition module 210 is further used to acquire a third image of the target obstacle after the obstruction ends; acquire feature points of the target obstacle in the third image; compare the number of feature points in the third image with the number of feature points in the first image; if the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold, determine that the target obstacle in the third image is the same as the target obstacle in the first image.

[0071] The prediction module 220 predicts the motion trajectory of the target obstacle according to the first image and the particle filter algorithm when the target obstacle is in an obstructed state.

[0072] Specifically, after the first image is sent to the controller, the controller can judge the target obstacle of the first image. When it is judged that the target obstacle is in an obscured state, the motion trajectory of the target obstacle can be predicted based on the first image and the particle filtering algorithm. The motion trajectory of the target obstacle may include parameters such as the position and speed of the target obstacle in multiple time frames.

[0073] In a possible implementation, the prediction module 220 is further used to obtain first state parameters and a preset tracking period of the target obstacle in the first image, where the first state parameters include a first position and a first speed; input the first state parameters and the preset tracking period into a preset state transfer equation to obtain second state parameters of the target obstacle, where the second state parameters include a second position and a second speed; and generate a motion trajectory of the target obstacle according to the first state parameters and the second state parameters.

[0074] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0075] Figure 3 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application.

[0076] See also Figure 3 , the electronic device 300 includes a memory 310 and a processor 320.

[0077] The processor 320 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0078] The memory 310 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 320 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 310 may include any combination of computer-readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 310 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0079] The memory 310 stores executable codes, and when the executable codes are processed by the processor 320 , the processor 320 can execute part or all of the methods described above.

[0080] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0081] Alternatively, the present application can also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium) on which executable code (or computer program or computer instruction code) is stored. When the executable code (or computer program or computer instruction code) is executed by a processor of an electronic device (or server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0082] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A target tracking method, characterized in that: include: Acquire a first image captured by a visual sensor of the vehicle during driving, wherein the first image includes a target obstacle; When the target obstacle is in an obstructed state, a motion trajectory of the target obstacle is predicted according to the first image and a particle filtering algorithm.

2. The method according to claim 1, characterized in that The predicting the motion trajectory of the target obstacle according to the first image and the particle filtering algorithm includes: Acquire a first state parameter and a preset tracking period of the target obstacle in the first image, wherein the first state parameter includes a first position and a first speed; Inputting the first state parameter and the preset tracking period into a preset state transfer equation to obtain a second state parameter of the target obstacle, wherein the second state parameter includes a second position and a second speed; A motion trajectory of the target obstacle is generated according to the first state parameter and the second state parameter.

3. The method according to claim 2, characterized in that The formula of the state transfer equation is as follows: in, xi ,y i are the horizontal and vertical coordinates of the second position, respectively, v x 、v y are the lateral speed of the second speed and the longitudinal speed of the second speed respectively, Δt is the preset tracking period, is the process noise, x i-1 ,y i-1 are the abscissa and ordinate of the first position, is the lateral speed of the first speed and the longitudinal speed of the first speed.

4. The method according to claim 1, characterized in that Also includes: Acquire a second image after the first image, and calculate a displacement between the first image and the second image; If the displacement is less than a second threshold, it is determined that the target obstacle is in an obstructed state; If the displacement is greater than or equal to the second threshold, it is determined that the target obstacle is not in an obstructed state.

5. The method according to claim 4, characterized in that The calculating the displacement between the first image and the second image comprises: Extracting feature points of the target obstacle from the first image and the second image respectively; The displacement between the two images is calculated based on the feature points of at least two images.

6. The method according to claim 5, characterized in that The extracting feature points of the target obstacle from the first image and the second image respectively includes: For any one of the first image and the second image, extract brightness data of all pixels of the target obstacle from the image, and take pixels whose brightness data difference with adjacent pixels is greater than a first threshold as target pixels; The target pixel point is compared with the characteristic pixel points of the target obstacle pre-stored in the database, so as to use the target pixel point matching the characteristic pixel point as the characteristic point of the target obstacle.

7. The method according to any one of claims 2 to 6, characterized in that Also includes: Acquire a third image of the target obstacle after the target obstacle stops being blocked; Acquire feature points of the target obstacle in the third image; comparing the number of feature points in the third image with the number of feature points in the first image; If the difference between the number of feature points in the third image and the number of feature points in the first image is less than a third threshold, it is determined that the target obstacle in the third image is the same as the target obstacle in the first image.

8. A target tracking device, characterized in that: include: An acquisition module is used to acquire a first image captured by a visual sensor during the driving process of the vehicle, wherein the first image includes a target obstacle; A prediction module predicts a motion trajectory of the target obstacle according to the first image and a particle filter algorithm when the target obstacle is in an obstructed state.

9. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: An executable code is stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as claimed in any one of claims 1 to 7.

Images