Method, system, medium and equipment for constructing visual perception feature descriptor

Abstract

The present invention discloses a method, system, medium and equipment for constructing a visual perception feature descriptor of ground objects. The method includes the following steps: obtaining ground object information within a preset range under a vehicle body; constructing a vehicle body coordinate system, and obtaining local feature coordinates based on the vehicle body coordinate system; calculating the Euclidean distance and angle between each ground object coordinate and the coordinates of other ground objects within a preset range under the vehicle body, and obtaining the spatial topological relationship vector of each ground object based on the Euclidean distance and angle corresponding to each ground object; constructing a ground object feature descriptor based on the ground object information, local feature coordinates and the spatial topological relationship vector of each ground object; combining the ground object semantic information, spatial position information coordinates and spatial topological relationship obtained by the perception system to increase the dimension of the perception data, thereby improving the availability of the perception data even in scenarios with insufficient perception accuracy and a lack of ground objects.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving systems, and in particular to a method, system, medium and device for constructing a visual perception ground feature descriptor. Background Art

[0002] The perception module in the intelligent driving system obtains the semantic features and spatial position information of objects and applies it to tasks such as navigation, positioning, planning and control. However, due to the influence of environmental factors and model accuracy, the category, shape, size, and spatial position of perceived objects may not all meet the requirements of the intelligent driving system. Especially in scenes where objects are not abundant, how to make full use of the information of perceived objects and improve data utilization is an urgent problem that needs to be solved to ensure the smooth completion of corresponding tasks in the intelligent driving system. Summary of the Invention

[0003] The present invention provides a method, system, medium and device for constructing a visual perception feature descriptor of land objects, which combines the semantic information of land objects, spatial position information coordinates and spatial topological relationships obtained by the perception system to increase the dimension of perception data. Even in scenarios with insufficient perception accuracy and a small number of land objects, the usability of perception data can be improved.

[0004] In a first aspect, a method for constructing a visually perceived object feature descriptor is provided, comprising the following steps:

[0005] Obtain information about ground objects within a preset range under the vehicle body;

[0006] Constructing a vehicle body coordinate system, and obtaining coordinates of various objects based on the vehicle body coordinate system;

[0007] Calculate the Euclidean distance and angle between each feature coordinate and the coordinates of other features within a preset range under the vehicle body, and obtain the spatial topological relationship vector of each feature based on the Euclidean distance and angle corresponding to each feature;

[0008] A feature descriptor is constructed based on the feature information, the coordinates of each feature and the spatial topological relationship vector of each feature.

[0009] According to the first aspect, in a first possible implementation of the first aspect, the step of “calculating the Euclidean distance and angle between each ground object coordinate and the coordinates of other ground objects within a preset range under the vehicle body” specifically includes the following steps:

[0010] Divide a preset range under the vehicle body into grids of preset sizes;

[0011] According to the coordinate value of each feature within the preset range under the vehicle body, the features are divided into corresponding grids one by one;

[0012] Calculate the Euclidean distance between each feature coordinate in the grid and the coordinates of other features in the grid;

[0013] Calculate the angle between each object coordinate in the grid and the coordinates of other objects in the grid.

[0014] According to the first possible implementation manner of the first aspect, in the second possible implementation manner of the first aspect, the step of “calculating the Euclidean distance between each object coordinate in the grid and the coordinates of other objects in the grid” specifically includes the following steps:

[0015] Assume that one of the object coordinates in the grid is the coordinate of the object to be measured;

[0016] When it is detected that there is no object in other cells of the grid, the Euclidean distance between the coordinates of the object to be detected and the cell where no object is detected is zero;

[0017] When an object is detected in another cell of the grid, the Euclidean distance between the coordinates of the object to be detected and the coordinates of an object in the cell where an object is detected is calculated;

[0018] When multiple objects are detected in other cells of the grid, the average Euclidean distance between the coordinates of the object to be detected and the coordinates of multiple objects in the cell where the multiple objects are detected is calculated.

[0019] According to the first possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the step of “calculating the angle between each object coordinate in the grid and the coordinates of other objects in the grid” specifically includes the following steps:

[0020] Assume that one of the object coordinates in the grid is the coordinate of the object to be measured;

[0021] When it is detected that there is no object in other cells of the grid, the angle between the coordinates of the object to be detected and the cell where no object is detected is zero;

[0022] When a ground object is detected in another cell of the grid, the angle formed by the coordinates of the ground object to be detected and the coordinates of a ground object in the cell where the ground object is detected in the vehicle coordinate system is calculated;

[0023] When multiple objects are detected in other cells of the grid, the center point coordinates of the multiple object coordinates in the cells where multiple objects are detected are calculated, and the angle formed by the coordinates of the object to be measured and the center point coordinates in the vehicle coordinate system is calculated.

[0024] According to the first aspect, in a fourth possible implementation of the first aspect, the step of “obtaining a spatial topological relationship vector of each feature based on the Euclidean distance and angle corresponding to each feature” specifically includes the following steps:

[0025] The Euclidean distance and angle corresponding to each feature are recorded in sequence into a preset length vector to obtain the spatial topological relationship vector of each feature.

[0026] According to the first aspect, in a fifth possible implementation of the first aspect, the step of “constructing a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature” specifically includes the following steps:

[0027] The feature information, feature coordinates and spatial topological relationship vector corresponding to each feature are connected in series to construct a feature feature descriptor.

[0028] In the second aspect, a system for constructing a visually perceived object feature descriptor is provided, including:

[0029] An acquisition module is used to obtain ground feature information within a preset range under the vehicle body;

[0030] A coordinate module, in communication with the acquisition module, for constructing a vehicle body coordinate system and obtaining coordinates of various objects based on the vehicle body coordinate system;

[0031] a calculation module, in communication with the coordinate module, for calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body, and obtaining a spatial topological relationship vector of each object based on the Euclidean distance and angle corresponding to each object; and

[0032] The feature descriptor construction module is in communication with the acquisition module, the coordinate module and the calculation module, and is used to construct a feature descriptor of a ground object according to the ground object information, the coordinates of each ground object and the spatial topological relationship vector of each ground object.

[0033] In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the method for constructing a visual perception feature descriptor is implemented as described above.

[0034] In a fourth aspect, an electronic device is provided, comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein the processor implements the method for constructing a visually perceived feature descriptor as described above when executing the computer program.

[0035] Compared with the existing technology, the advantages of the present invention are as follows: by combining the semantic information of land objects and spatial position information - land object coordinates obtained by the perception system, and by further mining to obtain the spatial topological relationship of land objects, the dimension of perception data is increased. Even in scenarios with insufficient perception accuracy and a small number of land objects, the semantics, spatial position, and global topological relationship of land objects can be fully utilized to improve the availability of perception data. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of an embodiment of a method for constructing a visually perceived object feature descriptor according to the present invention;

[0037] Figure 2 This is a flow chart of another embodiment of a method for constructing a visually perceived object feature descriptor according to the present invention;

[0038] Figure 3 It is a structural diagram of a visual perception ground feature descriptor construction system of the present invention. DETAILED DESCRIPTION

[0039] Reference will now be made in detail to specific embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Although the present invention will be described in conjunction with specific embodiments, it will be understood that the present invention is not intended to be limited to those embodiments. On the contrary, it is intended to cover variations, modifications, and equivalents within the spirit and scope of the present invention as defined by the appended claims. It should be noted that the method steps described herein can be implemented by any functional block or functional arrangement, and any functional block or functional arrangement can be implemented as a physical entity or a logical entity, or a combination of the two.

[0040] In order to enable those skilled in the art to better understand the present invention, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] Note: The following example is only a specific example and is not intended to limit the embodiments of the present invention to the following specific steps, values, conditions, data, sequence, etc. Those skilled in the art can apply the concepts of the present invention to construct more embodiments not described in this specification by reading this specification.

[0042] See also Figure 1 As shown, an embodiment of the present invention provides a method for constructing a visual perception feature descriptor, comprising the following steps:

[0043] S100, obtaining ground feature information within a preset range under the vehicle body;

[0044] S200, constructing a vehicle body coordinate system, and obtaining coordinates of various objects according to the vehicle body coordinate system;

[0045] S300, calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body, and obtaining a spatial topological relationship vector of each object based on the Euclidean distance and angle corresponding to each object;

[0046] S400: Constructing a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature.

[0047] Specifically, in this embodiment, S100 obtains real-time visually perceived ground object information under the vehicle body, retaining only the ground objects within the range of the vehicle body length M*width N under the vehicle system; extracts the color, category, shape, area and other information of each ground object within the range of M*N meters and stores it in the perceived ground object list.

[0048] S200, constructing a vehicle body coordinate system. The coordinate axis is centered on the vehicle body, the horizontal axis is perpendicular to the vehicle body's forward direction, and the vertical axis is parallel to the vehicle body's forward direction. The right side of the horizontal axis is positive, and the left side is negative. The front side of the vertical axis is positive, and the rear side is negative. The angles subsequently calculated start with the vehicle body's forward direction, with clockwise directions being positive and counterclockwise directions being negative.

[0049] The present invention proposes a method for constructing a feature descriptor of visually perceived objects. This method combines the semantic information of objects and the spatial position information (object coordinates) obtained by the perception system, and further mines the spatial topological relationship of objects to increase the dimension of perception data. Even in scenarios with insufficient perception accuracy and a small number of objects, it can fully utilize the semantics, spatial position, and global topological relationship of objects to improve the usability of perception data.

[0050] Preferably, in another embodiment of the present application, the step of "S300, calculating the Euclidean distance and angle between each feature coordinate and the coordinates of other features within a preset range under the vehicle body" specifically includes the following steps:

[0051] Divide a preset range under the vehicle body into grids of preset sizes;

[0052] According to the coordinate value of each feature within the preset range under the vehicle body, the features are divided into corresponding grids one by one;

[0053] Calculate the Euclidean distance between each feature coordinate in the grid and the coordinates of other features in the grid;

[0054] Calculate the angle between each object coordinate in the grid and the coordinates of other objects in the grid.

[0055] Specifically, in this embodiment, a spatial distribution matrix - grid of size M*N is constructed in the preset range under the vehicle body in the vehicle body coordinate system, and the matrix is ​​initially a 0 matrix; and the coordinates of each object in the vehicle body coordinate system are judged one by one. If the x coordinate of the object is between (m-1,m) and the y coordinate is between (n-1,n), then 1 is added to the mth row and nth column of the spatial distribution matrix; the value range of m is [1,M], and m can be any value therein, and the same applies to n.

[0056] The following table (1) shows the distribution location of each feature in the grid;

[0057]

[0058] Note: If 20 objects are detected within the preset range of the vehicle body, the 20 objects will be divided into corresponding grids one by one.

[0059] Preferably, in another embodiment of the present application, the step of "calculating the Euclidean distance between each feature coordinate in the grid and the coordinates of other features in the grid" specifically includes the following steps:

[0060] Assume that one of the object coordinates in the grid is the coordinate of the object to be measured;

[0061] When it is detected that there is no object in other cells of the grid, the Euclidean distance between the coordinates of the object to be detected and the cell where no object is detected is zero;

[0062] When an object is detected in another cell of the grid, the Euclidean distance between the coordinates of the object to be detected and the coordinates of an object in the cell where an object is detected is calculated;

[0063] When multiple objects are detected in other cells of the grid, the average Euclidean distance between the coordinates of the object to be detected and the coordinates of multiple objects in the cell where the multiple objects are detected is calculated.

[0064] Specifically, in this embodiment, referring to the above table (1), taking feature ① as an example, the difference between feature ① in the grid and feature ② in other grids is calculated. The Euclidean distance between them is as follows:

[0065] 1. There are two features ②③ in the second cell of the grid. dis1 is the Euclidean distance between ① and ②, and dis2 is the Euclidean distance between ① and ③. Then the average Euclidean distance is: dis = (dis1 + dis2) / 2.

[0066] 2. If there is no feature in the third cell of the grid, the Euclidean distance is 0.

[0067] 3. There is a feature ④ in the fourth cell of the grid, then dis is the Euclidean distance between ① and ④.

[0068] 4. Calculate the Euclidean distance between ① and the features in the remaining cells using the above calculation method.

[0069] Landform②... The corresponding Euclidean distance results are the same as the above steps and will not be repeated here.

[0070] It should be noted that the calculation formula for the Euclidean distance between two feature coordinates is as follows:

[0071] The coordinates of feature A are (x A ,y A), object B (the object B can be any other object in the vehicle coordinate system except object A) is (x B ,y B ), then the Euclidean distance between the two is

[0072] Preferably, in another embodiment of the present application, the step of "calculating the angle between each object coordinate in the grid and the coordinates of other objects in the grid" specifically includes the following steps:

[0073] Assume that one of the object coordinates in the grid is the coordinate of the object to be measured;

[0074] When it is detected that there is no object in other cells of the grid, the angle between the coordinates of the object to be detected and the cell where no object is detected is zero;

[0075] When a ground object is detected in another cell of the grid, the angle formed by the coordinates of the ground object to be detected and the coordinates of a ground object in the cell where the ground object is detected in the vehicle coordinate system is calculated;

[0076] When multiple objects are detected in other cells of the grid, the center point coordinates of the multiple object coordinates in the cells where multiple objects are detected are calculated, and the angle formed by the coordinates of the object to be measured and the center point coordinates in the vehicle coordinate system is calculated.

[0077] Specifically, in this embodiment, when multiple objects are detected in other cells of the grid, the center point coordinates of the multiple object coordinates in the cell where the multiple objects are detected are calculated as follows:

[0078]

[0079] Therefore, in the vehicle coordinate system defined above, the angle formed by the object A and the center point coordinates in the vehicle coordinate system can be obtained using the following formula:

[0080]

[0081] Among them, the coordinates of feature A are (x A ,y A ), the center point coordinates are (x c ,y c ).

[0082] Similarly, the ground feature ①... The corresponding angle result is consistent with the above Euclidean distance calculation steps and will not be repeated here.

[0083] Preferably, in another embodiment of the present application, the step of "S300, obtaining a spatial topological relationship vector of each feature according to the Euclidean distance and angle corresponding to each feature" specifically includes the following steps:

[0084] S330 , recording the Euclidean distance and angle corresponding to each feature in sequence into a preset length vector to obtain a spatial topological relationship vector of each feature.

[0085] Specifically, in this embodiment, the Euclidean distance and angle corresponding to each feature are sequentially recorded into a vector of length 2*M*N to obtain a spatial topological relationship vector of each feature.

[0086] Preferably, in another embodiment of the present application, the step of "S400, constructing a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature" specifically includes the following steps:

[0087] The feature information, feature coordinates and spatial topological relationship vector corresponding to each feature are connected in series to construct a feature feature descriptor.

[0088] Specifically, in this embodiment, the color, category, shape, area, coordinates in the vehicle coordinate system, and spatial topological relationship vector of the feature A are connected in series.

[0089] The color, category, shape, area, coordinates in the vehicle coordinate system and spatial topological relationship vector of feature B are connected in series.

[0090] Therefore, the feature descriptor can be constructed by concatenating the information of each feature in the above manner.

[0091] See also Figure 2 As shown, a method for constructing a visual perception feature descriptor provided by an embodiment of the present invention includes the following steps:

[0092] 1. Retain the features within the range of length M*width N of the vehicle body under the vehicle system, and extract the semantic information of the features within the range of length M*width N;

[0093] 2. Constructing a vehicle body coordinate system, and obtaining the coordinates of various objects (spatial position information) based on the vehicle body coordinate system;

[0094] 3. Construct an M*N spatial distribution matrix in the vehicle coordinate system and calculate the Euclidean distance and angle corresponding to each feature;

[0095] 4. Obtain the spatial topological relationship vector of each feature;

[0096] 5. Construct a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature.

[0097] See also Figure 3 As shown, an embodiment of the present invention provides a system for constructing a visually perceived object feature descriptor, including:

[0098] An acquisition module is used to obtain ground feature information within a preset range under the vehicle body;

[0099] A coordinate module, in communication with the acquisition module, for constructing a vehicle body coordinate system and obtaining coordinates of various objects based on the vehicle body coordinate system;

[0100] a calculation module, in communication with the coordinate module, for calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body, and obtaining a spatial topological relationship vector of each object based on the Euclidean distance and angle corresponding to each object; and

[0101] The feature descriptor construction module is in communication with the acquisition module, the coordinate module and the calculation module, and is used to construct a feature descriptor of a ground object according to the ground object information, the coordinates of each ground object and the spatial topological relationship vector of each ground object.

[0102] Therefore, the present invention proposes a visual perception feature descriptor construction system, which combines the feature semantic information and spatial position information (feature coordinates) obtained by the perception system, and obtains the spatial topological relationship of the features through further mining, thereby increasing the dimension of the perception data. Even in scenarios with insufficient perception accuracy and a small number of features, it can fully utilize the semantics, spatial position, and global topological relationship of the features to improve the availability of perception data.

[0103] Specifically, this embodiment corresponds one-to-one to the above method embodiment, and the functions of each module have been described in detail in the corresponding method embodiment, so they will not be repeated here.

[0104] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, all or part of the method steps of the above method are implemented.

[0105] The present invention implements all or part of the process in the above method, and can also be completed by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of each of the above method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0106] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program running on the processor, and when the processor executes the computer program, all or part of the method steps in the above method are implemented.

[0107] The processor may be a central processing unit (CPU), 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 any conventional processor. The processor is the control center of a computer device and connects various parts of the entire computer device using various interfaces and lines.

[0108] The memory can be used to store computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created based on the use of the mobile phone (such as audio data, video data, etc.). In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0109] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, servers, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage) containing computer-usable program code.

[0110] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), servers, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0111] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0113] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for constructing a visually perceived feature descriptor, characterized in that: The following steps are involved: Obtain information about ground objects within a preset range under the vehicle body; Constructing a vehicle body coordinate system, and obtaining coordinates of various objects based on the vehicle body coordinate system; Calculate the Euclidean distance and angle between each feature coordinate and the coordinates of other features within a preset range under the vehicle body, and obtain the spatial topological relationship vector of each feature based on the Euclidean distance and angle corresponding to each feature; Constructing a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature; The step of "calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body" specifically includes the following steps: Divide a preset range under the vehicle body into grids of preset sizes; According to the coordinate value of each feature within the preset range under the vehicle body, the features are divided into corresponding grids one by one; Calculate the Euclidean distance between each feature coordinate in the grid and the coordinates of other features in the grid; Calculate the angle between each object coordinate in the grid and the coordinates of other objects in the grid.

2. The method for constructing a visually perceived feature descriptor according to claim 1, wherein: The step of "calculating the Euclidean distance between each object coordinate in the grid and the coordinates of other objects in the grid" specifically includes the following steps: Assume that one of the object coordinates in the grid is the coordinate of the object to be measured; When it is detected that there is no object in other cells of the grid, the Euclidean distance between the coordinates of the object to be detected and the cell where no object is detected is zero; When an object is detected in another cell of the grid, the Euclidean distance between the coordinates of the object to be detected and the coordinates of an object in the cell where an object is detected is calculated; When multiple objects are detected in other cells of the grid, the average Euclidean distance between the coordinates of the object to be detected and the coordinates of multiple objects in the cell where the multiple objects are detected is calculated.

3. The method for constructing a visually perceived feature descriptor according to claim 1, wherein: The step of "calculating the angle between each object coordinate in the grid and the coordinates of other objects in the grid" specifically includes the following steps: Assume that one of the object coordinates in the grid is the coordinate of the object to be measured; When it is detected that there is no object in other cells of the grid, the angle between the coordinates of the object to be detected and the cell where no object is detected is zero; When a ground object is detected in another cell of the grid, the angle formed by the coordinates of the ground object to be detected and the coordinates of a ground object in the cell where the ground object is detected in the vehicle coordinate system is calculated; When multiple objects are detected in other cells of the grid, the center point coordinates of the multiple object coordinates in the cells where multiple objects are detected are calculated, and the angle formed by the coordinates of the object to be measured and the center point coordinates in the vehicle coordinate system is calculated.

4. The method for constructing a visually perceived feature descriptor according to claim 1, wherein: The step of "obtaining the spatial topological relationship vector of each feature according to the Euclidean distance and angle corresponding to each feature" specifically includes the following steps: The Euclidean distance and angle corresponding to each feature are recorded in sequence into a preset length vector to obtain the spatial topological relationship vector of each feature.

5. The method for constructing a visually perceived feature descriptor according to claim 1, wherein: The step of "constructing a feature descriptor based on the feature information, the coordinates of each feature, and the spatial topological relationship vector of each feature" specifically includes the following steps: The feature information, feature coordinates and spatial topological relationship vector corresponding to each feature are connected in series to construct a feature feature descriptor.

6. A visual perception feature descriptor construction system, characterized by: include: An acquisition module is used to obtain ground feature information within a preset range under the vehicle body; A coordinate module, in communication with the acquisition module, for constructing a vehicle body coordinate system and obtaining coordinates of various objects based on the vehicle body coordinate system; a calculation module, in communication with the coordinate module, for calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body, and obtaining a spatial topological relationship vector of each object based on the Euclidean distance and angle corresponding to each object; and A feature descriptor construction module, which is in communication with the acquisition module, the coordinate module and the calculation module, and is used to construct a feature descriptor based on the feature information, the coordinates of each feature and the spatial topological relationship vector of each feature; The step of "calculating the Euclidean distance and angle between each object coordinate and the coordinates of other objects within a preset range under the vehicle body" specifically includes the following steps: Divide a preset range under the vehicle body into grids of preset sizes; According to the coordinate value of each feature within the preset range under the vehicle body, the features are divided into corresponding grids one by one; Calculate the Euclidean distance between each feature coordinate in the grid and the coordinates of other features in the grid; Calculate the angle between each object coordinate in the grid and the coordinates of other objects in the grid.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for constructing a visually perceived object feature descriptor according to any one of claims 1 to 5 is implemented.

8. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor runs the computer program, the method for constructing a visually perceived object feature descriptor according to any one of claims 1 to 5 is implemented.

Images