Monocular perception data processing method and apparatus, electronic device, and storage medium

Abstract

Disclosed are a monocular perception data processing method and apparatus, an electronic device, and a storage medium. The monocular perception data processing method comprises: acquiring a monocular image collected by a monocular camera, the monocular image comprising at least one obstacle; converting the monocular image into point cloud data; converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system, the two-dimensional polar coordinate system using the monocular camera as an origin; dividing the two-dimensional polar coordinate system into a plurality of equally divided phase intervals, the plurality of phase intervals using the origin as a vertex; determining a target polar coordinate point in each phase interval, the target polar coordinate point being a coordinate point having the smallest radius among polar coordinate points in all intervals within the phase intervals; and determining position information of the obstacle on the basis of all of the target polar coordinate points.

Description

Monocular perception data processing method and device, electronic equipment and storage medium

[0001] Reference of Related Applications

[0002] The present disclosure claims the full right of priority of the Chinese Invention Patent Application No. 202411343475.0, filed on September 25, 2024, and entitled “Monocular perception data processing method and device, electronic equipment and storage medium”, and incorporates it by reference in its entirety.

[0003] TECHNICAL FIELD

[0004] The present disclosure generally relates to the technical field of data processing, and more particularly to a monocular perception data processing method and device, electronic equipment and storage medium.

[0005] BACKGROUND

[0006] Currently, a common low-cost configuration of intelligent autonomous mobile robots often uses a monocular camera as the main perception sensor to meet the basic obstacle detection, boundary detection, moving object detection and other perception needs in the application scenario. The common process is to obtain an image and then perform segmentation / detection and other inferences of an artificial intelligence (AI) model, and the result is converted into point cloud in a 3D space for obstacle avoidance and boundary restriction. Since the monocular camera lacks depth information, the implementation of this process relies on an important prior assumption: assuming that the obstacle is located on a horizontal ground.

[0007] Although this assumption is met in most cases in open outdoor scenarios, there are still many times when the obstacle does not meet the assumption, i.e., the obstacle itself has a certain height, such as a standing person, and objects in the scene with a certain height, such as trees, etc. In this case, the above process of converting image information into point cloud will produce obvious error estimation, resulting in the inability to accurately determine the position information of the obstacle, thereby interfering with the obstacle avoidance behavior of the robot.

[0008] SUMMARY

[0009] In a first aspect, the present disclosure provides a monocular perception data processing method, comprising:

[0010] obtaining a monocular image collected by a monocular camera; wherein the monocular image includes at least one obstacle;

[0011] converting the monocular image into point cloud data;

[0012] converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system; wherein the two-dimensional polar coordinate system takes the monocular camera as the origin;

[0013] The two-dimensional polar coordinate system is divided into multiple equal phase intervals; wherein the multiple phase intervals have the origin as a vertex;

[0014] A target polar coordinate point in each phase interval is determined; wherein the target polar coordinate point is the coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and

[0015] The position information of the obstacle is determined according to all the target polar coordinate points.

[0016] Optionally, the monocular image is converted into point cloud data, comprising:

[0017] Depth information of the monocular image is estimated;

[0018] An intrinsic matrix of the monocular camera is obtained;

[0019] The monocular image is converted into the point cloud data according to the depth information and the intrinsic matrix.

[0020] In some embodiments, the depth information of the monocular image is estimated, comprising:

[0021] The depth information of the monocular image is estimated by using a pre-trained depth estimation model;

[0022] Alternatively,

[0023] A first image captured by the monocular camera is obtained; wherein the first image has a different shooting angle and / or a different shooting position from the monocular image; and the depth information of the monocular image is estimated according to the first image and the monocular image.

[0024] In some embodiments, the point cloud data is converted into polar coordinate data in a two-dimensional polar coordinate system, comprising:

[0025] The point cloud data is projected onto a horizontal plane to obtain horizontal coordinate data on the horizontal plane;

[0026] The horizontal coordinate data is converted into a two-dimensional polar coordinate system to obtain radius data and angle data; and

[0027] The radius data and the angle data are taken as the polar coordinate data.

[0028] In some embodiments, the two-dimensional polar coordinate system is divided into multiple equal phase intervals, comprising:

[0029] A target number of division intervals is determined; and

[0030] The two-dimensional polar coordinate system is divided into the target number of equal phase intervals.

[0031] In some embodiments, determining the target number of division intervals comprises:

[0032] obtaining an accuracy requirement; and

[0033] determining the target number of division intervals according to the accuracy requirement and a preset mapping relationship; wherein, the mapping relationship is a mapping relationship between accuracy requirement and number of division intervals.

[0034] In some embodiments, determining a target polar coordinate point in each of the phase intervals comprises:

[0035] determining all interval polar coordinate points contained in the phase interval;

[0036] obtaining radius information of the interval polar coordinate points;

[0037] determining target radius information with the smallest radius among all the radius information; and

[0038] taking the interval polar coordinate point corresponding to the target radius information as the target polar coordinate point of the phase interval.

[0039] In a second aspect, the present disclosure provides a monocular perception data processing device, the device comprising:

[0040] an acquisition module configured to acquire a monocular image collected by a monocular camera; wherein the monocular image comprises at least one obstacle;

[0041] a first conversion module configured to convert the monocular image into point cloud data;

[0042] a second conversion module configured to convert the point cloud data into polar coordinate data in a two-dimensional polar coordinate system; wherein the two-dimensional polar coordinate system takes the monocular camera as an origin;

[0043] a coordinate system division module configured to divide the two-dimensional polar coordinate system into a plurality of equally divided phase intervals; wherein the plurality of phase intervals take the origin as a vertex;

[0044] a first determination module configured to determine a target polar coordinate point in each of the phase intervals; wherein the target polar coordinate point is a coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and

[0045] a second determination module configured to determine position information of the obstacle according to all the target polar coordinate points.

[0046] In a third aspect, the present disclosure provides an electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus.

[0047] a memory configured to store a computer program;

[0048] a processor configured to implement the monocular perception data processing method of the present disclosure when executing the computer program stored in the memory.

[0049] In a fourth aspect, the present disclosure provides a computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the monocular perception data processing method of the present disclosure.

[0050] In some embodiments, the point cloud data obtained based on the monocular image is converted into polar coordinate data in a two-dimensional polar coordinate system, by screening the coordinate points divided into each phase interval, the coordinate point with the smallest radius is retained as the target polar coordinate point, the downsampling of the point cloud data is realized, the error data in the process of converting the monocular image into point cloud data is eliminated, and the position information of the obstacle is determined based on the target polar coordinate point retained in each phase interval, thereby improving the identification accuracy of the position information of the obstacle.

[0051] BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated herein and constitute part of the specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor.

[0054] One or more embodiments are exemplarily illustrated by the pictures in the drawings corresponding thereto, and these exemplary illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, unless otherwise specified. The drawings in the drawings do not constitute a proportional limitation.

[0055] FIG. 1 is a system architecture diagram of a monocular perception data processing method according to an embodiment of the present disclosure;

[0056] FIG. 2 is a flowchart of a monocular perception data processing method according to an embodiment of the present disclosure;

[0057] FIG. 3 is a schematic diagram of converting a monocular image into point cloud data;

[0058] FIG. 4 is a schematic diagram of dividing a two-dimensional polar coordinate system according to an embodiment of the present disclosure;

[0059] FIG. 5 is a structural schematic diagram of a monocular perception data processing device according to an embodiment of the present disclosure; and

[0060] FIG. 6 is a structural schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0061] To make the objects, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are some but not all of the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present disclosure.

[0062] The following disclosure provides many different embodiments, or examples, for implementing different structures of the present disclosure. For the purpose of simplicity and clarity, the description of the specific examples in the following text will be described. Of course, they are only examples and the purpose is not to limit the present disclosure. In addition, reference numerals and / or letters can be repeated in different examples in the present disclosure. Such repetition is for the purpose of simplification and clarity, and does not indicate the relationship between the various embodiments and / or settings being discussed.

[0063] The present disclosure provides a monocular perception data processing method, device, electronic device, and storage medium, which can eliminate error data in the process of converting monocular images into point cloud data, determine the position information of obstacles based on the target polar coordinate points retained in each phase interval, and improve the identification accuracy of the position information of obstacles.

[0064] The first embodiment of the present disclosure provides a monocular perception data processing method, which can be applied to a system architecture as shown in FIG. 1, the system architecture at least includes a data acquisition module 101 and a data processing module 102, and the data acquisition module 101 and the data processing module 102 are in communication connection. In some embodiments, the system architecture can be a smart autonomous mobile robot equipped with a monocular camera, such as a lawn mower robot, a sweeping robot, etc., without limitation.

[0065] In some embodiments, based on the system architecture, the monocular perception data processing method is described in detail, as shown in FIG. 2, the monocular perception data processing method includes:

[0066] Step 201, obtaining a monocular image collected by a monocular camera; wherein the monocular image includes at least one obstacle;

[0067] Step 202, converting the monocular image into point cloud data;

[0068] Step 203, converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system; wherein the two-dimensional polar coordinate system takes the monocular camera as the origin;

[0069] Step 204, dividing the two-dimensional polar coordinate system into multiple equal phase intervals; wherein the multiple phase intervals take the origin as the vertex.

[0070] Step 205, determining the target polar coordinate point in each phase interval; wherein the target polar coordinate point is the coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and

[0071] Step 206, determining the position information of the obstacle according to all target polar coordinate points.

[0072] In some embodiments, the intelligent autonomous mobile robot is equipped with a monocular camera, and the monocular camera is used to identify obstacles in the working scene. When the intelligent autonomous mobile robot is working, the monocular camera is used to collect monocular images.

[0073] In some embodiments, converting the monocular image into point cloud data includes: estimating the depth information of the monocular image; obtaining the intrinsic matrix of the monocular camera; and converting the monocular image into point cloud data according to the depth information and the intrinsic matrix.

[0074] In some embodiments, since the monocular image lacks depth information, the intelligent autonomous mobile robot first estimates the depth information of the collected monocular image, and then converts the monocular image into point cloud data with depth information in combination with the intrinsic matrix of the monocular camera.

[0075] In some embodiments, the depth information of the monocular image can be estimated in at least two ways:

[0076] The first way is to use a pre-trained depth estimation model to estimate the depth information of the monocular image.

[0077] The depth estimation model can be trained using monocular training samples, so that after obtaining the monocular image, the depth information of the monocular image can be estimated using the depth estimation model.

[0078] The second way is to obtain a first image collected by the monocular camera; wherein the first image has a different shooting angle and / or a different shooting position from the monocular image; and estimate the depth information of the monocular image according to the first image and the monocular image.

[0079] The depth information of one of the monocular images can be determined using multiple monocular images obtained by the monocular camera at different shooting angles or different shooting positions.

[0080] Of course, the point cloud data obtained by the above method is not accurate because it is estimated from a monocular image, and only the distance information of the person or object in the image can be roughly estimated, and the height information cannot be determined. If the point cloud is projected onto the horizontal plane based on this, when the obstacle has a certain height (such as a standing person, a tree trunk, a table, a chair, etc.), the projected point cloud is only accurate at the camera (corresponding to the part of the obstacle close to the ground), and most of the point cloud is the result of incorrect projection (corresponding to the part of the obstacle not close to the ground and having a height). A schematic diagram of converting a monocular image into point cloud data is shown in FIG. 3, wherein a is a monocular image captured by a monocular camera, when a standing obstacle is converted into an obstacle point cloud, the point cloud is distributed on a horizontal ground as shown in b of FIG. 3, wherein the viewing angle of b of FIG. 3 is the same as that of a of FIG. 3, and c of FIG. 3 is a top view of the point cloud data, that is, from the perspective of looking down on the horizontal ground, the obstacle point cloud converted from the standing obstacle is actually "lying" on a horizontal ground.

[0081] In some embodiments, the point cloud data is converted into polar coordinate data in a two-dimensional polar coordinate system, including: projecting the point cloud data onto a horizontal plane to obtain horizontal coordinate data on the horizontal plane; converting the horizontal coordinate data into the two-dimensional polar coordinate system to obtain radius data and angle data; and taking the radius data and the angle data as the polar coordinate data.

[0082] In some embodiments, in a camera coordinate system, the coordinates of any point in the original point cloud are p(x, y, z), and z is ignored (because z does not affect navigation), and the point p is transformed into p'(r, theta) in a two-dimensional polar coordinate system with the monocular camera as the origin, where r represents the radius of the point p' to the origin, and theta represents the angle of the point p' in the two-dimensional polar coordinate system. The position of the point p' in the two-dimensional polar coordinate system is represented by the radius data and the angle data.

[0083] In some embodiments, the two-dimensional polar coordinate system is divided into a plurality of equal phase intervals, including: determining a target number of intervals; and dividing the two-dimensional polar coordinate system into the target number of equal phase intervals.

[0084] In some embodiments, the two-dimensional polar coordinate system can be divided into a plurality of equal phase intervals, such as dividing the 180° range in front of the camera into 180 intervals with the origin as the vertex, of course, it can also be divided into 90, 45 or 360, etc. The number of interval divisions is not limited. The 180° behind the camera can not be divided or can be divided but not considered, which is not limited.

[0085] In some embodiments, the target number of the divided intervals can be a preconfigured fixed number or can be determined according to the accuracy requirement.

[0086] In some embodiments, determining the target number of the divided intervals includes: obtaining the accuracy requirement; and determining the target number of the divided intervals according to the accuracy requirement and a preset mapping relationship, where the mapping relationship is a mapping relationship between the accuracy requirement and the number of the divided intervals.

[0087] In some embodiments, the target number of the divided intervals can be determined according to the obtained accuracy requirement and the preset mapping relationship between the accuracy requirement and the number of the divided intervals, so that the number of the divided intervals is flexibly adjusted according to the accuracy requirement, and the number of the divided intervals is ensured to meet the accuracy requirement of the user. When the accuracy requirement is high, a larger number of the divided intervals can be selected, and when the accuracy requirement is low, a smaller number of the divided intervals can be selected. The number of the divided intervals is not specifically limited in the embodiments of the present disclosure, but is determined according to the accuracy requirement, so that the accuracy is considered while ensuring the speed of data processing, and high accuracy or low accuracy is not blindly selected.

[0088] In some embodiments, a two-dimensional polar coordinate system division schematic diagram is shown in FIG. 4. The two-dimensional polar coordinate system can be divided into a plurality of same phase intervals. For example, a 180° range in front of the camera with the origin as the vertex can be divided into a plurality of intervals with the same angle. A 180° range behind the camera can not be divided because no image is collected. The angle of each phase interval is the same. The polar coordinate points contained in the phase interval can be down-sampled to eliminate error data in the process of converting the monocular image into point cloud data, so that the position information of the obstacle can be determined based on the target polar coordinate points retained in each phase interval, and the identification accuracy of the position information of the obstacle is improved.

[0089] It should be noted that the number of the divided intervals in FIG. 4 only represents an example of how to divide, and does not represent a limitation on the number of the divided intervals.

[0090] In some embodiments, determining the target polar coordinate point in each phase interval includes: determining all interval polar coordinate points contained in the phase interval; obtaining radius information of the interval polar coordinate points; determining target radius information with the smallest radius among all the radius information; and taking the interval polar coordinate point corresponding to the target radius information as the target polar coordinate point of the phase interval.

[0091] In some embodiments, referring to FIG. 4, different phase intervals contain different numbers of interval polar coordinate points. For any one phase interval, the coordinate point with the smallest radius among all the interval polar coordinate points in the phase interval can be taken as the target polar coordinate point. That is, only one coordinate point is reserved for each phase interval containing interval polar coordinate points, so as to realize the downsampling processing of the polar coordinate points contained in each phase interval. It should be understood that the phase intervals not containing interval polar coordinate points can not be processed. Since in the process of converting monocular images into point clouds, the point cloud data obtained by projection is only accurate point cloud (corresponding to the part of the obstacle close to the ground) near the camera, and most of the point cloud is the result of incorrect projection (corresponding to the part of the obstacle not close to the ground and having a height). Therefore, reserving the polar coordinate point with the smallest radius can retain accurate point cloud data and eliminate inaccurate point cloud data. That is, the point cloud processed by the method of the embodiment of the present disclosure not only retains the core obstacle information (the point cloud mapped by the bottom of the obstacle, which has accurate obstacle position information), but also automatically filters out the point cloud obtained by incorrect projection due to the height of the obstacle. At the same time, the number of point clouds is also reduced, which is helpful for reducing the communication bandwidth pressure between the perception and navigation modules in the robot system and reducing the calculation amount of the navigation module on the obstacle cost map for the intelligent autonomous mobile robot applying the method.

[0092] In some embodiments, in view of the important prior assumption "assuming that the obstacle is located on the horizontal ground" relied on in the common process of the current monocular perception, when dealing with the problem of perceiving obstacles with a certain height, the downsampling algorithm considering occlusion is performed on the point cloud converted to the polar coordinate system, so as to realize the retention of the core obstacle information and the filtering of the point cloud obtained by incorrect projection due to the height of the obstacle, while reducing the number of point clouds without affecting obstacle avoidance and boundary limitation, and reducing the system data transmission and calculation load.

[0093] In some embodiments, the point cloud data obtained based on the monocular image is converted into polar coordinate data in a two-dimensional polar coordinate system. By screening the coordinate points divided into each phase interval and reserving the coordinate point with the smallest radius as the target polar coordinate point, the downsampling of the point cloud data is realized, the incorrect data in the process of converting the monocular image into point cloud data is eliminated, the position information of the obstacle is determined based on the target polar coordinate point reserved in each phase interval, and the identification accuracy of the position information of the obstacle is improved.

[0094] Based on the same technical concept, the second embodiment of the present disclosure provides a monocular perception data processing device, as shown in FIG. 5, which comprises:

[0095] The acquisition module 501 is configured to acquire a monocular image collected by a monocular camera; wherein the monocular image comprises at least one obstacle.

[0096] The first conversion module 502 is configured to convert the monocular image into point cloud data.

[0097] The second conversion module 503 is configured to convert the point cloud data into polar coordinate data in a two-dimensional polar coordinate system, wherein the two-dimensional polar coordinate system takes the monocular camera as an origin.

[0098] The coordinate system division module 504 is configured to divide the two-dimensional polar coordinate system into a plurality of equally divided phase intervals, wherein the plurality of phase intervals take the origin as a vertex.

[0099] The first determination module 505 is configured to determine a target polar coordinate point in each phase interval, wherein the target polar coordinate point is a coordinate point with the smallest radius among all interval polar coordinate points in the phase interval.

[0100] The second determination module 506 is configured to determine the position information of the obstacle according to all target polar coordinate points.

[0101] In some embodiments, the point cloud data obtained based on the monocular image is converted into polar coordinate data in a two-dimensional polar coordinate system, by screening the coordinate points divided into each phase interval and retaining the coordinate point with the smallest radius as the target polar coordinate point, the downsampling of the point cloud data is realized, the error data in the process of converting the monocular image into point cloud data is eliminated, the position information of the obstacle is determined based on the target polar coordinate point retained in each phase interval, and the identification accuracy of the position information of the obstacle is improved.

[0102] As shown in FIG. 6, the electronic device provided by the embodiments of the present disclosure includes a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112 and the memory 113 complete mutual communication through the communication bus 114,

[0103] The memory 113 is configured to store a computer program.

[0104] The processor 111 is configured to execute the program stored in the memory 113 to implement a monocular perception data processing method, wherein the method includes:

[0105] obtaining a monocular image collected by a monocular camera, wherein the monocular image includes at least one obstacle;

[0106] converting the monocular image into point cloud data;

[0107] converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system, wherein the two-dimensional polar coordinate system takes the monocular camera as an origin.

[0108] The two-dimensional polar coordinate system is divided into a plurality of equal phase intervals, wherein the plurality of phase intervals have the origin as a vertex;

[0109] A target polar coordinate point in each phase interval is determined, wherein the target polar coordinate point is a coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and

[0110] The position information of the obstacle is determined according to all the target polar coordinate points.

[0111] In some embodiments, the communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0112] The communication interface is configured to communicate between the terminal and other devices.

[0113] In some embodiments, the memory can include a Random Access Memory (RAM) and can also include a non-volatile memory, such as at least one disk memory. Optionally, the memory can also be at least one storage device located away from the aforementioned processor.

[0114] In some embodiments, the processor mentioned above can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; can also be a Digital Signal Processing (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0115] The embodiments of the present disclosure further provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the monocular perception data processing method provided by the present disclosure.

[0116] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0117] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software plus a general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0118] It should be understood that the terms used herein are for the purpose of describing specific example embodiments only and are not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises", "comprising", "includes", "including" and "has" are to be construed as containing, comprising, including or having, as set forth herein, and therefore should be interpreted symmetrically. The method steps, processes, and operations described herein are not to be interpreted as necessarily requiring their performance in the specific order described or illustrated, unless explicitly stated otherwise. It should also be understood that additional or alternative steps can be employed.

[0119] It should be understood that the specific embodiments described herein merely serve to explain the present disclosure and are not intended to limit the present disclosure. In the description, the suffixes such as "module", "part", or "unit" used to indicate elements are merely for convenience of explanation of the present disclosure, and have no specific meaning by themselves. Therefore, "module", "part", or "unit" can be used interchangeably.

[0120] The foregoing is merely illustrative of the various implementations of the present disclosure and the general principles thereof. Numerous modifications can be made to these illustrations, and equivalents can be substituted therefor, without departing from the scope of the present disclosure. The specific embodiments commensurate with the specific application are intended to be illustrative only and not limiting of the scope of the application as set forth in the following claims.

Claims

1. A method for processing monocular perception data, comprising: obtaining a monocular image captured by a monocular camera, wherein the monocular image comprises at least one obstacle; converting the monocular image into point cloud data; converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system, wherein the two-dimensional polar coordinate system takes the monocular camera as an origin; dividing the two-dimensional polar coordinate system into a plurality of equally divided phase intervals, wherein the plurality of phase intervals take the origin as a vertex; determining a target polar coordinate point in each phase interval, wherein the target polar coordinate point is a coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and determining position information of the obstacle according to all target polar coordinate points. 2.The method of claim 1, wherein converting the monocular image into point cloud data comprises: estimating depth information of the monocular image; obtaining an intrinsic matrix of the monocular camera; and converting the monocular image into the point cloud data according to the depth information and the intrinsic matrix. 3.The method of claim 2, wherein estimating the depth information of the monocular image comprises: estimating the depth information of the monocular image by using a pre-trained depth estimation model; or obtaining a first image captured by the monocular camera, wherein the first image has a different shooting angle and / or a different shooting position from the monocular image; and estimating the depth information of the monocular image according to the first image and the monocular image. 4.The method of any one of claims 1 to 3, wherein converting the point cloud data into polar coordinate data in a two-dimensional polar coordinate system comprises: projecting the point cloud data onto a horizontal plane to obtain horizontal coordinate data on the horizontal plane; converting the horizontal coordinate data to a two-dimensional polar coordinate system to obtain radius data and angle data; and taking the radius data and the angle data as the polar coordinate data. 5.The method of any one of claims 1 to 4, wherein dividing the two-dimensional polar coordinate system into a plurality of equally divided phase intervals comprises: determining a target number of divided intervals; and dividing the two-dimensional polar coordinate system into the target number of equally divided phase intervals. 6.The method of claim 5, wherein determining the target number of divided intervals comprises: obtaining a precision requirement; and determining the target number of divided intervals according to the precision requirement and a preset mapping relationship, wherein the mapping relationship is a mapping relationship between a precision requirement and a number of divided intervals. 7.The method of any one of claims 1 to 6, wherein determining a target polar coordinate point in each phase interval comprises: determining all interval polar coordinate points contained in the phase interval; obtaining radius information of the interval polar coordinate points; determining target radius information with the smallest radius among all the radius information; and taking the interval polar coordinate point corresponding to the target radius information as the target polar coordinate point of the phase interval. 8.A device for processing monocular perception data, comprising: ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ An acquisition module configured to acquire a monocular image collected by a monocular camera; wherein the monocular image comprises at least one obstacle; A first conversion module configured to convert the monocular image into point cloud data; A second conversion module configured to convert the point cloud data into polar coordinate data in a two-dimensional polar coordinate system; wherein the two-dimensional polar coordinate system takes the monocular camera as an origin; A coordinate system division module configured to divide the two-dimensional polar coordinate system into a plurality of equally divided phase intervals; wherein the plurality of phase intervals take the origin as a vertex; A first determination module configured to determine a target polar coordinate point in each phase interval; wherein the target polar coordinate point is a coordinate point with the smallest radius among all interval polar coordinate points in the phase interval; and A second determination module configured to determine position information of the obstacle according to all target polar coordinate points.

9. An electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein, The processor, the communication interface, and the memory complete mutual communication through the communication bus; The memory is configured to store a computer program; The processor is configured to execute the program stored on the memory to implement the monocular perception data processing method in any one of claims 1 to 7.

10. A computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the monocular perception data processing method in any one of claims 1 to 7.

Images