Target tracking method and device, electronic equipment and storage medium

Abstract

The invention provides a target tracking method and device, electronic equipment and a storage medium, and relates to the technical field of intelligent driving. The method comprises the steps that when a target vehicle enters the recognition range of a first camera of the vehicle, the target vehicle is recognized through the first camera to determine feature points of the target vehicle; tracking the feature points to determine image coordinates when the feature points leave the recognition range of the first camera; and according to the image coordinates and an alignment function between the first camera and the second camera, determining a position interval of the image of the target vehicle appearing in the second camera, and according to the position interval, tracking the target vehicle. According to the invention, the two-dimensional image information feature of the area where the target vehicle is located in the image of the first camera can be converted into one-dimensional information, so that the computing power consumption is reduced; and when the target is tracked in the second camera subsequently, the target vehicle can be tracked directly based on the position interval, so that the computing power consumption is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a target tracking method, device, electronic device and storage medium. Background Art

[0002] Intelligent driving (also known as automatic driving or unmanned driving) refers to a driving mode in which the vehicle can automatically perceive the environment, plan routes, control driving, and cope with complex road conditions with no human intervention or a small amount of human intervention; intelligent driving technology relies on the collaborative work of multiple sensors, algorithms and computing platforms. Vehicles use multiple sensors to perceive the surrounding environment. For example, cameras can provide visual information and identify targets such as lane lines, vehicles, and pedestrians. LiDAR can generate high-precision three-dimensional environmental point clouds through laser ranging to detect the distance and shape of the target. Millimeter-wave radar can detect the speed and distance of long-distance targets. Ultrasonic sensors can achieve close-range detection at low speeds, and then make decisions based on the perception and positioning information to control the speed and direction of the vehicle and achieve operations such as changing lanes, overtaking, parking or deceleration.

[0003] However, since intelligent driving systems need to process large amounts of image data in real time in complex environments and perform tasks such as precise feature extraction, target recognition, and path planning, target tracking consumes a lot of computing power, which not only increases the cost of the hardware platform, but also increases the computing pressure. Summary of the invention

[0004] The problem solved by the present invention is how to reduce the computing power consumption of target tracking.

[0005] To solve the above problems, the present invention provides a target tracking method, device, electronic device and storage medium.

[0006] In a first aspect, the present invention provides a target tracking method, comprising:

[0007] When a target vehicle enters the recognition range of the first camera of the vehicle, the target vehicle is recognized by the first camera to determine the characteristic points of the target vehicle;

[0008] Tracking the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera;

[0009] The position interval in which the target vehicle appears in the image of the second camera is determined according to the image coordinates and an alignment function between the first camera and the second camera, and the target vehicle is tracked according to the position interval.

[0010] Optionally, determining a position interval where the target vehicle appears in the image of the second camera according to the image coordinates and an alignment function between the first camera and the second camera comprises:

[0011] Determine the speed of the target vehicle according to the running length of the feature point in the image of the second camera;

[0012] Determine, according to the speed of the target vehicle, a time interval from when the target vehicle leaves the recognition range of the first camera to when it enters the recognition range of the second camera, and determine, according to the time interval, the alignment function corresponding to the speed of the target vehicle;

[0013] The position interval is determined according to the image coordinates and the alignment function.

[0014] Optionally, determining the speed of the target vehicle according to the running length of the feature point in the image of the second camera includes:

[0015] Sampling the running track of the feature point in the image of the second camera multiple times to determine the sampling point of the feature point;

[0016] Determine the vector between the sampling points according to the sampling points at different times, and determine the component of the feature point in the vehicle travel direction according to the vector;

[0017] Determine the running length according to the components, and determine the relative speed between the target vehicle and the host vehicle according to the running length and a vehicle speed calibration matrix;

[0018] The speed of the target vehicle is determined according to the relative vehicle speed and the speed of the host vehicle.

[0019] Optionally, determining the component of the feature point in the vehicle traveling direction according to the vector includes:

[0020] Fitting the sampling points to determine a fitting function;

[0021] The component of the feature point in the vehicle traveling direction is determined according to the fitting function.

[0022] Optionally, the vehicle speed calibration matrix is ​​configured to be calibrated in the following manner:

[0023] Continuously tracking the calibration feature points of the calibration vehicle in the image of the first camera and the image of the second camera at different relative vehicle speeds to determine the running trajectory of the calibration feature points, wherein the relative vehicle speed is the relative vehicle speed between the calibration vehicle and the host vehicle;

[0024] Determine, according to the running trajectory, a movement range of the calibrated feature point in a specific period of time in the image of the first camera and the image of the second camera, and determine, according to the movement range, a change distance of the calibrated feature point at different relative vehicle speeds;

[0025] The vehicle speed calibration matrix is ​​determined according to the change distance of the calibration feature points and the corresponding relative vehicle speed.

[0026] Optionally, tracking the target vehicle according to the location interval includes:

[0027] After determining the position interval, taking the time after the time interval from the time when the feature point leaves the recognition range of the first camera as the tracking time;

[0028] At the tracking moment, an area corresponding to the position interval in the image of the second camera is identified to track the target vehicle.

[0029] Optionally, the alignment function is configured to be calibrated in the following manner:

[0030] Repeating the process of calibrating the vehicle multiple times so that the vehicle leaves the recognition range of the first camera and enters the recognition range of the second camera, and determining a first position and a first time when the calibrated vehicle leaves the recognition range of the first camera each time;

[0031] Determining a second position and a second time each time the calibration vehicle enters a recognition range of the second camera;

[0032] Determining a range in which the calibration vehicle appears in the image of the second camera according to the second positions determined multiple times;

[0033] Determining a calibration time interval from when the calibrated vehicle leaves a recognition range of the first camera to when it enters a recognition range of the second camera according to the first moment and the second moment;

[0034] The alignment function is determined according to the first position, the range in which the calibration vehicle appears in the image of the second camera, and the calibration time interval.

[0035] In a second aspect, the present invention provides a target tracking device, comprising:

[0036] The first module is used to identify the target vehicle through the first camera when the target vehicle enters the recognition range of the first camera of the vehicle to determine the characteristic points of the target vehicle;

[0037] A second module is used to track the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera;

[0038] The third module is used to determine the position interval where the target vehicle appears in the image of the second camera according to the image coordinates and the alignment function between the first camera and the second camera, and track the target vehicle according to the position interval.

[0039] In a third aspect, the present invention provides an electronic device, including a memory and a processor;

[0040] The memory is used to store computer programs;

[0041] The processor is used to implement the target tracking method as described in the first aspect when executing the computer program.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the target tracking method as described in the first aspect is implemented.

[0043] The beneficial effect of the target tracking method of the present invention is: when the target vehicle enters the recognition range of the first camera of the vehicle, the target vehicle is identified by the first camera to determine the feature points of the target vehicle, that is, the two-dimensional image information features of the area where the target vehicle is located in the image of the first camera are converted into one-dimensional information, thereby reducing the amount of target data and thus reducing computing power consumption; then the feature points are tracked to determine the image coordinates of the feature points when they leave the recognition range of the first camera, and finally the position interval in the image of the second camera where the target vehicle appears can be determined based on the image coordinates. When the target is subsequently tracked in the second camera, the target vehicle can be tracked directly based on the position interval without searching in the entire image, thereby effectively reducing computing power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A schematic diagram of a target tracking method according to an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a process for determining a location interval according to an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of a process for determining the speed of a target vehicle according to an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of a flow chart of determining a component of a feature point in a vehicle traveling direction according to an embodiment of the present invention;

[0048] Figure 5A schematic diagram of a calibration process of a vehicle speed calibration matrix according to an embodiment of the present invention;

[0049] Figure 6 A schematic diagram of a process for tracking a target vehicle according to an embodiment of the present invention;

[0050] Figure 7 A schematic diagram of a calibration process of an alignment function according to an embodiment of the present invention;

[0051] Figure 8 A system architecture diagram of a target tracking device according to an embodiment of the present invention;

[0052] Fig. 9 4 is a system architecture diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0054] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0055] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0056] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0057] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.

[0058] like Figure 1 As shown, an embodiment of the present invention provides a target tracking method, comprising:

[0059] S100: When a target vehicle enters a recognition range of a first camera of the vehicle, the target vehicle is recognized by the first camera to determine feature points of the target vehicle.

[0060] Specifically, taking overtaking by this vehicle as an example, when this vehicle approaches the target vehicle from the right rear and enters a certain area behind the target vehicle, the target vehicle enters the recognition range of the first camera of this vehicle (taking the front-view camera as an example), and the front-view camera captures the target in the picture. The forward binocular system can achieve ranging of the target vehicle, and the edge center point (i.e., feature point) on the target vehicle is located through target recognition.

[0061] Among them, when the vehicle is overtaken by the target vehicle, the first camera can be a rear-view camera.

[0062] S200: Track the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera.

[0063] Specifically, the target vehicle is simulated as a feature point for continuous tracking, and the image coordinates of the feature point when it leaves the recognition range of the front-view camera can be determined.

[0064] If there are multiple target vehicles, a parallel processing strategy is adopted to continuously track the feature points of each target vehicle to obtain the motion trajectory of each target vehicle.

[0065] S300: Determine a position interval in which the target vehicle appears in the image of the second camera according to the image coordinates and an alignment function between the first camera and the second camera, and track the target vehicle according to the position interval.

[0066] Specifically, the position interval in which the target vehicle appears in the image of the side view camera is determined based on the image coordinates when the feature point leaves the recognition range of the front view camera and the alignment function between the first camera and the second camera (taking the side view camera as an example). When the target is subsequently tracked in the side view camera, the target vehicle can be tracked directly based on the position interval without searching in the entire image, which effectively reduces computing power consumption.

[0067] Among them, in vehicles with L2+ and higher levels of intelligent driving capabilities, multiple cameras will be configured (for example, at least 12), including front-view, rear-view, side-view and other cameras, and many cameras will work together during driving.

[0068] In this embodiment, when the target vehicle enters the recognition range of the first camera of this vehicle, the target vehicle is identified by the first camera to determine the feature points of the target vehicle, that is, the two-dimensional image information features of the area where the target vehicle is located in the image of the first camera are converted into one-dimensional information, thereby reducing the amount of target data and thus reducing computing power consumption; the feature points are then tracked to determine the image coordinates of the feature points when they leave the recognition range of the first camera, and finally, the position interval in which the target vehicle appears in the image of the second camera can be determined based on the image coordinates. When the target is subsequently tracked in the second camera, the target vehicle can be tracked directly based on the position interval without searching in the entire image, effectively reducing computing power consumption.

[0069] Optionally, determining a position interval where the target vehicle appears in the image of the second camera according to the image coordinates and an alignment function between the first camera and the second camera comprises:

[0070] S310: Determine the speed of the target vehicle according to the running length of the feature point in the image of the second camera.

[0071] Specifically, combined Figure 2 As shown, after sampling the feature points multiple times, the running length of the feature points in the image of the second camera is determined, and then the relative speed of the target vehicle and the host vehicle is obtained in combination with the speed calibration matrix, and then the speed of the target vehicle is obtained in combination with the host vehicle speed.

[0072] S320: Determine a time interval from when the target vehicle leaves the recognition range of the first camera to when it enters the recognition range of the second camera according to the speed of the target vehicle, and determine the alignment function corresponding to the speed of the target vehicle according to the time interval.

[0073] Specifically, combined Figure 2 As shown, the alignment function F(y)=Y[Y,Y n ,t](t∈T 间隔 ), at different target vehicle speeds, the corresponding time intervals are different, so the corresponding alignment functions are also different, and the corresponding alignment function is selected to determine the position interval.

[0074] In the alignment function, y refers to the second term in the position (x, y) when the target vehicle leaves the recognition range of the first camera.

[0075] S330: Determine the position interval according to the image coordinates and the alignment function.

[0076] Specifically, combined Figure 2 As shown, after determining the alignment functions corresponding to the speeds of different target vehicles, the position interval where the target vehicle appears in the image of the second camera can be determined in combination with the image coordinates.

[0077] In this optional embodiment, the alignment function corresponding to the target vehicle is determined according to the time interval, thereby determining the position interval, thereby improving the accuracy of the position interval.

[0078] Optionally, determining the speed of the target vehicle according to the running length of the feature point in the image of the second camera includes:

[0079] S311: Sampling the running trajectory of the feature point in the image of the second camera for multiple times to determine the sampling points of the feature point.

[0080] Specifically, combined Figure 3 As shown in the figure, when the target vehicle drives into the side of the vehicle, it enters the recognition range of the second camera of the vehicle (taking the side view camera as an example). For example, in the image of the side view camera, the target vehicle is at time T0, and the corresponding feature point is located at (X0, Y0). After a fixed time T A After that, the corresponding feature point is located at (X A , Y A ), and again after a fixed time T B After that, the corresponding feature point is located at (X B , Y B ), store the coordinates at the above time points; calculate at T A Within the time range, the running length L1 of the target vehicle in the image of the side view camera, and the running length L2 of the target vehicle in the image of the side view camera in the time range T B Within the time range, the running length L2 of the target vehicle in the image of the side view camera; in order to improve the accuracy of the vehicle speed estimation, multiple time samplings can be performed at unequal time intervals to determine the sampling points of the feature points.

[0081] S312: Determine a vector between the sampling points according to the sampling points at different times, and determine a component of the feature point in the vehicle moving direction according to the vector.

[0082] Specifically, combined Figure 3 As shown, the vector L between each sampling point is calculated by sampling points at different time points. 矢 , according to the vector L 矢 Determine the component L of the feature point in the direction of vehicle travel 纵 .

[0083] S313: Determine the running length according to the components, and determine the relative speed between the target vehicle and the host vehicle according to the running length and a vehicle speed calibration matrix.

[0084] Specifically, combined Figure 3 As shown, by increasing the number of sampling times t at equal intervals, the component L of the feature point in the direction of vehicle travel is 纵 The average calculation is performed to obtain the running length L, and then combined with the speed calibration matrix to obtain the relative speed between the target vehicle and the vehicle.

[0085] S314: Determine the speed of the target vehicle according to the relative vehicle speed and the speed of the host vehicle.

[0086] Specifically, combined Figure 3 As shown, the speed of the target vehicle is determined according to the relative speed and the speed of the host vehicle. For example, if the relative speed is 20 km / h and the speed of the host vehicle is 80 km / h, the speed of the target vehicle may be 6020 km / h.

[0087] In this optional embodiment, the speed of the target vehicle is determined by sampling the feature points, thereby improving the accuracy of the speed of the target vehicle and thus improving the accuracy of the position interval.

[0088] Optionally, determining the component of the feature point in the vehicle traveling direction according to the vector includes:

[0089] S3121: Fit the sampling points to determine a fitting function.

[0090] Specifically, combined Figure 4 As shown, due to L 矢 Contains the vehicle's L in the direction of travel 纵 and the L in the horizontal direction of the lane 横 Therefore, it is necessary to first use the least squares method to fit the function and obtain the fitting function F(x,y) to minimize the driving vibration (for example, the vehicle suddenly deviates to the left or right). As the time points increase, the vehicle's driving trajectory prediction will be more accurate, and then the F(x,y) vector value in each distance interval will be calculated.

[0091] S3122: Determine the component of the feature point in the vehicle travel direction according to the fitting function.

[0092] Specifically, combined Figure 4 As shown, the component L of the feature point in the vehicle's moving direction is determined according to the function direction of the fitting function. 纵 .

[0093] In this optional embodiment, the fitting function is determined by fitting the sampling points, and the component of the feature point in the vehicle's traveling direction is determined based on the fitting function, thereby improving the accuracy of the target vehicle's speed and thus improving the accuracy of the position interval.

[0094] Optionally, the vehicle speed calibration matrix is ​​configured to be calibrated in the following manner:

[0095] S3131: Continuously track the calibration feature points of the calibrated vehicle in the image of the first camera and the image of the second camera at different relative speeds to determine the running trajectory of the calibration feature points, wherein the relative speed is the relative speed between the calibrated vehicle and the host vehicle.

[0096] Specifically, combined Figure 5 As shown in the figure, taking the calibration vehicle in the second lane on the right and the vehicle in the first lane on the right as an example, when the calibration vehicle appears in the image of the front camera of the vehicle, the calibration vehicle is identified, and the center point of the upper edge of the vehicle target is selected as the target feature (hereinafter referred to as the feature point) for tracking. The vehicle moves forward at a constant speed and starts to overtake on the right, surpassing the calibration vehicle and driving to a certain distance in front of the calibration vehicle, and the overtaking process ends; during this process, the calibration vehicle appears in the field of view of the front camera and the side camera of the vehicle successively; by continuously tracking the calibration feature points, the running trajectory of the calibration feature points based on the fixed vehicle speed can be obtained.

[0097] S3132: Determine the movement range of the calibrated feature points in a specific period of time in the image of the first camera and the image of the second camera according to the running trajectory, and determine the change distance of the calibrated feature points at different relative vehicle speeds according to the movement range.

[0098] Specifically, combined Figure 5 As shown, based on marking at different distances, the movement range of the calibrated feature points in the image of the front view camera and the image of the side view camera during a specific period of time can be obtained, so that the change distance of the calibrated feature points under different relative vehicle speeds can be determined.

[0099] S3133: Determine the vehicle speed calibration matrix according to the change distance of the calibration feature point and the corresponding relative vehicle speed.

[0100] Specifically, combined Figure 5 As shown, by gradient adjusting the overtaking speed of the vehicle, approaching the calibration vehicle at different relative speed differences, and repeating the above steps at each gradient speed, the change distance of the calibration feature points at a certain relative speed can be obtained (between fixed road targets, such as between certain ground markings), so that the speed calibration matrix can be determined in combination with the relative speed of the calibration vehicle and the vehicle.

[0101] In this optional embodiment, by tracking the calibration feature points of the calibration vehicle and then determining the vehicle speed calibration matrix, the accuracy of the speed of the target vehicle is improved, thereby improving the accuracy of the position interval.

[0102] Optionally, tracking the target vehicle according to the location interval includes:

[0103] S340: After determining the position interval, the time after the time interval from the time when the feature point leaves the recognition range of the first camera is used as the tracking time.

[0104] Specifically, combined Figure 6 As shown, after the position interval is determined, timing starts from the moment when the feature point leaves the recognition range of the first camera, and after a time interval, the tracking moment is reached.

[0105] S350: At the tracking moment, an area corresponding to the position interval in the image of the second camera is identified to track the target vehicle.

[0106] Specifically, combined Figure 6 As shown, at the tracking moment, the area corresponding to the position interval in the image of the second camera is identified, so as to realize the tracking of the target vehicle. Since the tracking is not started until the tracking moment, the computing power consumption is effectively reduced.

[0107] In this optional embodiment, by waiting for a time interval from the moment when the feature point leaves the recognition range of the first camera to the tracking moment before starting tracking, the computing power consumption is effectively reduced.

[0108] Optionally, the alignment function is configured to be calibrated in the following manner:

[0109] S321: Repeat the process of calibrating the vehicle to drive out of the recognition range of the first camera and enter the recognition range of the second camera multiple times, and determine the first position and the first time when the calibrated vehicle leaves the recognition range of the first camera each time.

[0110] Specifically, combined Figure 7 As shown, taking the case where the calibration vehicle is stationary and the vehicle is traveling at a constant speed and overtaking the calibration vehicle as an example, the calibration vehicle drives out of the recognition range of the first camera and enters the recognition range of the second camera, and the first position and the first moment of the calibration vehicle leaving the recognition range of the first camera are determined each time; the position of the calibration vehicle when it drives out of the screen of the front-view camera is defined as (x1, y1, t1), wherein x1 and y1 are the coordinates of the calibration vehicle in the front-view screen, and t1 is the time of driving out.

[0111] S322: Determine a second position and a second time each time the calibrated vehicle enters the recognition range of the second camera.

[0112] Specifically, combined Figure 7 As shown, the position where the calibration vehicle appears in the side view camera's picture is defined as (X1, Y1, t2), where X1 and Y1 are the coordinates of the calibration vehicle in the side view picture, and t2 is the entry time.

[0113] S323: Determine a range in which the calibrated vehicle appears in the image of the second camera according to the second positions determined multiple times.

[0114] Specifically, combined Figure 7 As shown in the figure, by letting the calibration vehicle approach the left or right side of its lane respectively and repeating the above process for visual sampling, the range [Y, Y] where the calibration vehicle appears in the image of the side view camera can be obtained. n ].

[0115] S324: Determine a calibration time interval from when the calibrated vehicle leaves the recognition range of the first camera to when it enters the recognition range of the second camera according to the first moment and the second moment.

[0116] Specifically, combined Figure 7 As shown, when the vehicle is traveling at a constant speed, a calibration time interval from when the vehicle leaves the recognition range of the longitudinal camera to when it enters the recognition range of the side view camera is obtained according to the first moment and the second moment.

[0117] Among them, the calibration time interval can be expressed as: T 间隔 =t2-t1+t x , where t x It is the system alignment time, which is used to indicate that the calibrated vehicle appears out of sync in the two cameras due to differences in camera parameters, different physical positions between cameras, relative speed, and system judgment time.

[0118] S325: Determine the alignment function according to the first position, the range in which the calibrated vehicle appears in the image of the second camera, and the calibration time interval.

[0119] Specifically, combined Figure 7 As shown, according to the range of the calibrated vehicle in the image of the side view camera and T 间隔 Determine the alignment function, that is, set the alignment function F(y) = Y[Y,Y n ,t](t∈T 间隔 ).

[0120] In this optional embodiment, by calibrating the first camera and the second camera, it is ensured that the image data between the first camera and the second camera can work better together.

[0121] like Figure 8 As shown, an embodiment of the present invention provides a target tracking device 800, including:

[0122] The first module 810 is used to identify the target vehicle through the first camera to determine the feature points of the target vehicle when the target vehicle enters the recognition range of the first camera of the vehicle;

[0123] The second module 820 is used to track the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera;

[0124] The third module 830 is used to determine the position interval where the target vehicle appears in the image of the second camera according to the image coordinates and the alignment function between the first camera and the second camera, and track the target vehicle according to the position interval.

[0125] like Fig. 9 As shown, an electronic device 900 provided by an embodiment of the present invention includes a memory 920 and a processor 910; the memory 920 is used to store a computer program; the processor 910 is used to implement the target tracking method as described above when executing the computer program.

[0126] In other words, an electronic device 900 includes a memory 920 and a processor 910 coupled to the memory 920; the memory 920 is configured to store a computer program; and the processor 910 is configured to perform the following operations when executing the computer program:

[0127] When a target vehicle enters the recognition range of the first camera of the vehicle, the target vehicle is recognized by the first camera to determine the characteristic points of the target vehicle;

[0128] Tracking the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera;

[0129] The position interval in which the target vehicle appears in the image of the second camera is determined according to the image coordinates and an alignment function between the first camera and the second camera, and the target vehicle is tracked according to the position interval.

[0130] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the target tracking method as described above is implemented.

[0131] In other words, a non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following operations:

[0132] When a target vehicle enters the recognition range of the first camera of the vehicle, the target vehicle is recognized by the first camera to determine the characteristic points of the target vehicle;

[0133] Tracking the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera;

[0134] The position interval in which the target vehicle appears in the image of the second camera is determined according to the image coordinates and an alignment function between the first camera and the second camera, and the target vehicle is tracked according to the position interval.

[0135] An electronic device 900 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 900 is intended to represent various forms of digital electronic computer equipment, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 900 can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present invention described herein and / or required.

[0136] Electronic device 900 includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In RAM, various programs and data required for device operation can also be stored. The computing unit, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0137] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0138] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A target tracking method, characterized in that: include: When a target vehicle enters the recognition range of the first camera of the vehicle, the target vehicle is recognized by the first camera to determine the characteristic points of the target vehicle; Tracking the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera; The position interval in which the target vehicle appears in the image of the second camera is determined according to the image coordinates and an alignment function between the first camera and the second camera, and the target vehicle is tracked according to the position interval.

2. The target tracking method according to claim 1, characterized in that: Determining the position interval where the target vehicle appears in the image of the second camera according to the image coordinates and the alignment function between the first camera and the second camera comprises: Determine the speed of the target vehicle according to the running length of the feature point in the image of the second camera; Determine, according to the speed of the target vehicle, a time interval from when the target vehicle leaves the recognition range of the first camera to when it enters the recognition range of the second camera, and determine, according to the time interval, the alignment function corresponding to the speed of the target vehicle; The position interval is determined according to the image coordinates and the alignment function.

3. The target tracking method according to claim 2, characterized in that: Determining the speed of the target vehicle according to the running length of the feature point in the image of the second camera comprises: Sampling the running track of the feature point in the image of the second camera multiple times to determine the sampling point of the feature point; Determine the vector between the sampling points according to the sampling points at different times, and determine the component of the feature point in the vehicle travel direction according to the vector; Determine the running length according to the components, and determine the relative speed between the target vehicle and the host vehicle according to the running length and a vehicle speed calibration matrix; The speed of the target vehicle is determined according to the relative vehicle speed and the speed of the host vehicle.

4. The target tracking method according to claim 3, characterized in that: Determining the component of the feature point in the vehicle traveling direction according to the vector comprises: Fitting the sampling points to determine a fitting function; The component of the feature point in the vehicle traveling direction is determined according to the fitting function.

5. The target tracking method according to claim 3, characterized in that: The vehicle speed calibration matrix is ​​configured to be calibrated in the following manner: Continuously tracking the calibration feature points of the calibration vehicle in the image of the first camera and the image of the second camera at different relative vehicle speeds to determine the running trajectory of the calibration feature points, wherein the relative vehicle speed is the relative vehicle speed between the calibration vehicle and the host vehicle; Determine, according to the running trajectory, a movement range of the calibrated feature point in a specific period of time in the image of the first camera and the image of the second camera, and determine, according to the movement range, a change distance of the calibrated feature point at different relative vehicle speeds; The vehicle speed calibration matrix is ​​determined according to the change distance of the calibration feature points and the corresponding relative vehicle speed.

6. The target tracking method according to claim 2, characterized in that: The tracking of the target vehicle according to the position interval includes: After determining the position interval, taking the time after the time interval from the time when the feature point leaves the recognition range of the first camera as the tracking time; At the tracking moment, an area corresponding to the position interval in the image of the second camera is identified to track the target vehicle.

7. The target tracking method according to any one of claims 1 to 6, characterized in that: The alignment function is configured to be calibrated in the following manner: Repeating the process of calibrating the vehicle to drive out of the recognition range of the first camera and enter the recognition range of the second camera multiple times, and determining a first position and a first time when the calibrated vehicle leaves the recognition range of the first camera each time; Determining a second position and a second time each time the calibration vehicle enters a recognition range of the second camera; Determining a range in which the calibration vehicle appears in the image of the second camera according to the second positions determined multiple times; Determining a calibration time interval from when the calibrated vehicle leaves a recognition range of the first camera to when it enters a recognition range of the second camera according to the first moment and the second moment; The alignment function is determined according to the first position, the range in which the calibration vehicle appears in the image of the second camera, and the calibration time interval.

8. A target tracking device, characterized in that: include: The first module is used to identify the target vehicle through the first camera when the target vehicle enters the recognition range of the first camera of the vehicle to determine the characteristic points of the target vehicle; A second module is used to track the feature point to determine the image coordinates of the feature point when it leaves the recognition range of the first camera; The third module is used to determine the position interval where the target vehicle appears in the image of the second camera according to the image coordinates and the alignment function between the first camera and the second camera, and track the target vehicle according to the position interval.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the target tracking method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the target tracking method according to any one of claims 1 to 7 is implemented.

Images