Method, device, apparatus and computer-readable storage medium for evaluating map quality

Abstract

Embodiments of the present disclosure relate to methods, devices, and computer-readable storage media for evaluating map quality. A method for evaluating map quality includes identifying flat areas in a point cloud map based on semantic information of the point cloud map; and evaluating the quality of the point cloud map by evaluating at least one attribute of the flat area. Embodiments of the present disclosure can evaluate the quality of a point cloud map by analyzing the attributes of a flat area (e.g., a road surface area, etc.) in a point cloud map without comparing the point cloud map with a reference map, thereby enabling automated evaluation of the quality of the point cloud map.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to the field of maps, and more particularly to methods, devices, apparatuses, and computer-readable storage media for evaluating map quality. Background Art

[0002] High-precision point cloud maps are usually used in application scenarios such as object detection and high-precision positioning in autonomous driving. The quality of high-precision point cloud maps will significantly affect the accuracy of object detection and high-precision positioning. Therefore, it is particularly important to evaluate the quality of high-precision point cloud maps. Summary of the invention

[0003] Embodiments of the present disclosure provide a method, an apparatus, a device, and a computer-readable storage medium for evaluating map quality.

[0004] In a first aspect of the present disclosure, a method for evaluating map quality is provided, comprising: identifying a flat area in a point cloud map based on semantic information of the point cloud map; and evaluating the quality of the point cloud map by evaluating at least one attribute of the flat area.

[0005] In a second aspect of the present disclosure, a device for evaluating map quality is provided. The device comprises: a flat area recognition module configured to identify a flat area in the point cloud map based on semantic information of the point cloud map; and a quality assessment module configured to assess the quality of the point cloud map by assessing at least one attribute of the flat area.

[0006] In a third aspect of the present disclosure, an electronic device is provided, comprising one or more processors; and a memory for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the method according to the first aspect of the present disclosure.

[0007] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, any step of the method described in the first aspect of the present disclosure is implemented.

[0008] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the present disclosure, nor is it intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present disclosure.

[0010] Figure 1 A block diagram illustrating an example environment in which embodiments of the present disclosure can be implemented;

[0011] Figure 2 A schematic block diagram of a map evaluation device according to an embodiment of the present disclosure is shown;

[0012] Figure 3 A flowchart illustrating an example method for evaluating map quality according to an embodiment of the present disclosure is shown;

[0013] Figure 4 A flowchart illustrating an example method for evaluating smoothness of a flat area according to an embodiment of the present disclosure is shown;

[0014] Figure 5 A block diagram showing an example apparatus for evaluating map quality according to an embodiment of the present disclosure; and

[0015] Figure 6 A block diagram of an example electronic device capable of implementing various embodiments of the present disclosure is shown.

[0016] In the various drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0018] As used herein, the term "including" and its variations mean open inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0019] As mentioned above, high-precision point cloud maps are often used in application scenarios such as object detection and high-precision positioning in autonomous driving. The quality of high-precision point cloud maps will significantly affect the accuracy of object detection and high-precision positioning. Therefore, it is particularly important to evaluate the quality of high-precision point cloud maps. Some traditional solutions evaluate the quality of point cloud maps by comparing them with reference maps. However, the reference map may not be easy to obtain or the reference map itself may be inaccurate, which will affect the quality assessment of the point cloud map.

[0020] The embodiments of the present disclosure propose a solution for evaluating map quality, which can solve one or more of the above-mentioned problems and other potential problems. In this solution, based on the semantic information of the point cloud map, the flat areas in the point cloud map are identified. The quality of the point cloud map is evaluated by evaluating at least one attribute of the flat area. In this way, the solution can evaluate the quality of the point cloud map by analyzing the attributes of the flat areas (e.g., road surface areas, etc.) in the point cloud map without comparing the point cloud map with a reference map, thereby enabling automated evaluation of the quality of the point cloud map.

[0021] Figure 1 1 shows a block diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. Figure 1 As shown, environment 100 includes a map collection device 110 and a map evaluation device 120. It should be understood that the structure and function of environment 100 are described for exemplary purposes only, and do not imply any limitation on the scope of the present disclosure. For example, embodiments of the present disclosure may also be applied to environments different from environment 100.

[0022] The map acquisition device 110 may include, but is not limited to, a collection vehicle or other equipment for collecting map data. For example, a laser radar may be installed on the map acquisition device 110 for collecting data. The map acquisition device 110 may move within a specific geographic area during a collection period to collect point cloud data for making maps. The "point cloud data" described herein may refer to data information of each point on the surface of an object returned when a beam of laser is irradiated on the surface of the object, including the three-dimensional coordinates of each point (e.g., x-coordinate, y-coordinate, and z-coordinate) and the laser reflection intensity (also referred to as "reflection value"). The map acquisition device 110 may generate a point cloud map 115 about a specific geographic area based on the collected point cloud data.

[0023] The point cloud map 115 may be provided to a map evaluation device 120 . The map evaluation device 120 may evaluate the quality of the point cloud map 115 by analyzing the attributes of the road surface area in the point cloud map 115 to generate an evaluation result 125 .

[0024] Figure 2FIG. 1 shows a schematic block diagram of a map evaluation device 120 according to an embodiment of the present disclosure. Figure 2 As shown, the map evaluation device 120 may include a projection module 210, a flat area identification module 220, and an evaluation module 230. It should be understood that the structure and function of the map evaluation device 120 are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure. In some embodiments, the map evaluation device 120 may be used in conjunction with Figure 2 The structure shown can be implemented in different structures.

[0025] In some embodiments, the projection module 210 is configured to generate a projection map 215 corresponding to the point cloud map 115 by performing a two-dimensional grid projection on the point cloud map 115. The projection map 215 can be divided into multiple grids (also referred to as "multiple unit areas"). For each grid, the height mean and height variance corresponding to the grid can be determined based on the height value of the original coordinate point corresponding to the grid in the point cloud map 115. In addition, the reflection value mean and reflection value variance corresponding to the grid can be determined based on the reflection value of the original coordinate point corresponding to the grid in the point cloud map 115. In this article, the projection map 215 is also referred to as a "two-dimensional projection map" or a "two-dimensional grid map."

[0026] In some embodiments, in order to generate a projection map 215 corresponding to the point cloud map 115, the projection module 210 may first aggregate the coordinate points of the three-dimensional point cloud map 115 according to the three-dimensional grid, wherein each three-dimensional grid is a cube with a predetermined side length. For each three-dimensional grid, the projection module 210 may calculate the weighted results of all the original coordinate points falling therein, including the weighted results of the height value and the weighted results of the reflection value. The projection module 210 may count the mean height value, the variance of the height value, the mean reflection value and the variance of the reflection value for each three-dimensional grid. After obtaining the statistical results of each three-dimensional grid, the projection module 210 may divide the three-dimensional grid into a plurality of grid sets, wherein each grid set has the same horizontal coordinate (i.e., each grid set corresponds to the same x-coordinate and the same y-coordinate), and each grid set is also referred to as a "pillar" in this article. The projection module 210 may perform Euclidean clustering on the points in each pillar, thereby obtaining a plurality of point clusters. The projection module 210 may calculate the weighted results of the height value and the weighted results of the reflection value for each point cluster. The projection module 210 can filter out the aerial point clusters, dynamic point clusters, and point clusters with lower weights (e.g., clusters with fewer points) in each column. The projection module 210 can calculate the height value mean, height value variance, reflection value mean, and reflection value variance for each filtered column, thereby obtaining a projection map 215. For example, the plane area corresponding to each column is a unit area in the projection map 215.

[0027] The generation of the projection map 215 is described above for exemplary purposes only. It should be understood that in some embodiments, the projection module 210 may generate the projection map 215 corresponding to the point cloud map 115 in any other manner. The scope of the present disclosure is not limited in this respect.

[0028] In some embodiments, the flat area identification module 220 is configured to identify the flat area 225 in the projection map 215 based on the semantic information of the flat area. Examples of flat areas may include, but are not limited to, road surface areas, such as lanes, sidewalks, squares, etc. The semantic information of the flat area may be predetermined. Additionally or alternatively, the semantic information of the flat area may be updated accordingly based on at least a portion of the evaluation results 125, thereby improving its accuracy.

[0029] In some embodiments, the evaluation module 230 is configured to evaluate the quality of the point cloud map 115 by evaluating at least one attribute of the flat area 225, thereby generating the evaluation result 125. The at least one attribute may include at least one of the smoothness and thickness of the flat area. Figure 2 As shown, for example, the evaluation module 230 may include a smoothness evaluation module 231 and / or a thickness evaluation module 232 , wherein the smoothness evaluation module 231 is configured to evaluate the smoothness of the flat region 225 , and the thickness evaluation module 232 is configured to evaluate the thickness of the flat region 225 .

[0030] The smoothness of the flat area 225 is mainly reflected by the height change of the flat area 225. In some embodiments, in order to evaluate the smoothness of the flat area 225, the smoothness evaluation module 231 can obtain a pose graph corresponding to the point cloud map 115, and the pose graph includes a pose set of the map acquisition device 110 when acquiring the point cloud map 115. In some embodiments, for each unit area in a plurality of unit areas, the smoothness evaluation module 231 can determine at least one pose associated with the unit area from the pose graph. For example, the smoothness evaluation module 231 can determine at least one pose within the vicinity of the unit area. The determination of the vicinity of the unit area can be implemented based on a K-nearest neighbor (KNN) algorithm or other similar algorithms, and the scope of the present disclosure is not limited in this respect. In some embodiments, the smoothness evaluation module 231 can determine the relative height value of the unit area based on the height mean of the unit area in the projection map 215 and the corresponding height value of at least one pose. For example, the smoothness evaluation module 231 can perform a weighted average on the corresponding height value of at least one pose to obtain a height weighted average of at least one pose. The weight of the height value of each of the at least one pose may be determined based on the following formula:

[0031]

[0032] where x p,i represents the x coordinate of the ith pose, x cell Indicates the x-coordinate and y-coordinate of the unit area. p,i represents the y coordinate of the i-th pose and y cell Indicates the y coordinate of the unit area. MAX_WEIGHT indicates a preset maximum weight, for example, which is less than 1. The smoothness evaluation module 231 can calculate the difference between the height mean of the unit area in the projection map 215 and the weighted average of the heights of at least one posture as the relative height value of the unit area.

[0033] In some embodiments, the smoothness evaluation module 231 may generate a grayscale image by mapping multiple relative height values ​​of multiple unit areas into grayscale values ​​(for example, ranging from 0 to 255). The smoothness evaluation module 231 may generate a gradient map of the grayscale image by performing edge detection on the grayscale image. For example, edge detection may be implemented based on the Canny algorithm or other similar algorithms, and the scope of the present disclosure is not limited in this regard. Additionally or alternatively, the smoothness evaluation module 231 may binarize the gradient map, where a unit area with a value of 0 represents an area where the height change is lower than a predetermined threshold (also referred to as the "first threshold" herein), and a unit area with a value not equal to 0 represents an area where the height change exceeds a predetermined threshold (also referred to as the "first threshold" herein). The smoothness evaluation module 231 may determine, for each unit area in at least one unit area included in the flat area 225, whether the height change of the unit area exceeds the first threshold (that is, whether there is a gradient value not equal to 0 in the binarized gradient map). If a certain unit area has a gradient value that is not 0 in the binary gradient map, the map evaluation device 120 may determine the unit area as a smoothness outlier in the flat area 225 .

[0034] Additionally or alternatively, in some embodiments, the smoothness evaluation module 231 may further determine whether the distance between the smoothness abnormal point in the flat area 225 and the boundary of the flat area 225 is less than a predetermined threshold (also referred to herein as a "second threshold"). If the smoothness abnormal point is located near the boundary of the flat area 225 (that is, the distance from the boundary of the flat area 225 is lower than the second threshold), the smoothness evaluation module 231 may identify the smoothness abnormal point as a smoothness alarm point within the flat area 225, which indicates that the smoothness abnormality may be caused by an error in the semantic information of the flat area. If the smoothness abnormal point is not near the boundary of the flat area 225 (that is, the distance from the boundary of the flat area 225 exceeds the second threshold), the smoothness evaluation module 231 may identify the smoothness abnormal point as a smoothness error point within the flat area 225, which indicates that the smoothness abnormality may be caused by an error in the point cloud map. In some embodiments, the smoothness evaluation module 231 may generate statistical information related to the smoothness alarm points and / or the smoothness error points in the flat area 225 as a smoothness evaluation result. For example, the statistical information may indicate whether there are smoothness alarm points and / or smoothness error points in the flat area 225, their specific locations and proportions, etc.

[0035] The smoothness of the flat area 225 is mainly reflected by the range of height variation (i.e., height variance) of the flat area 225. In some embodiments, in order to evaluate whether the thickness of the flat area 225 is abnormal, the thickness evaluation module 232 can generate statistical results based on the corresponding height variance of at least one unit area included in the flat area 225. The statistical results may include the mean, maximum value, minimum value, etc. of the height variance. The thickness evaluation module 232 can determine whether the thickness of the flat area 225 is abnormal based on the statistical results and generate a thickness evaluation result.

[0036] Return to reference Figure 2 , the evaluation module 230 can generate the evaluation result 125 based on at least one of the smoothness evaluation result generated by the smoothness evaluation module 231 and the thickness evaluation result generated by the thickness evaluation module 232. In this way, the embodiments of the present disclosure can evaluate the quality of the point cloud map by analyzing the properties of the flat area (e.g., road surface area, etc.) in the point cloud map without comparing the point cloud map with the reference map, thereby enabling automatic evaluation of the quality of the point cloud map. The evaluation result 125 can be used to update the point cloud map 115 and / or update the map semantic information used to identify the flat area, thereby improving the quality of the point cloud map 115 and improving the accuracy of the map semantic information.

[0037] Figure 3FIG. 3 is a flowchart of an exemplary method 300 for evaluating map quality according to an embodiment of the present disclosure. The method 300 may be performed, for example, in Figure 1 and Figure 2 The map evaluation device 120 shown is executed. Figure 2 The method 300 is described in detail. It should be understood that the method 300 may also include blocks not shown and / or may omit blocks shown. The scope of the present disclosure is not limited in this respect.

[0038] At block 310 , the map evaluation device 120 identifies the flat areas 225 in the point cloud map 115 based on the semantic information of the point cloud map 115 .

[0039] In some embodiments, in order to identify the flat area 225, the map evaluation device 120 may obtain a projection map 215 corresponding to the point cloud map 115 by performing a two-dimensional grid projection on the point cloud map 115, wherein the projection map 215 is divided into a plurality of unit areas, each unit area having a height mean and a height variance determined based on the height value of the original coordinate point corresponding to the unit area in the point cloud map 115. The map evaluation device 120 may identify the flat area 225 in the projection map 215 based on the semantic information, wherein the flat area 225 includes at least one unit area among the plurality of unit areas.

[0040] At block 320 , the map evaluation device 120 evaluates the quality of the point cloud map by evaluating at least one property of the flat area.

[0041] In some embodiments, the map evaluation device 120 may evaluate the quality of the point cloud map 115 by evaluating at least one of the smoothness and thickness of the flat area 225 .

[0042] In some embodiments, to evaluate the thickness of the flat area 225 , the map evaluation device 120 may generate statistical results based on the corresponding height variance of at least one unit area, and then determine whether the thickness of the flat area 225 is abnormal based on the statistical results.

[0043] In some embodiments, to evaluate the smoothness of the flat area 225 , the map evaluation device 120 may determine smoothness anomaly points, smoothness warning points, and / or smoothness error points within the flat area 225 , and generate relevant statistical information.

[0044] Figure 4 4 is a flowchart of an example method 400 for evaluating the smoothness of a flat area according to an embodiment of the present disclosure. The method 400 can be viewed as an example implementation of block 330, which can be implemented in, for example, Figure 1 and Figure 2The method 400 is executed at the map evaluation device 120 shown. It should be understood that the method 400 may also include blocks not shown and / or may omit blocks shown. The scope of the present disclosure is not limited in this respect.

[0045] like Figure 4 As shown, at box 410 , the map evaluation device 120 may obtain a pose graph corresponding to the point cloud map 115 , where the pose graph includes a pose set of the map acquisition device 110 when acquiring the point cloud map 115 .

[0046] At block 420 , the map evaluation device 120 determines a plurality of relative height values ​​for a plurality of unit areas based on the projection map 215 and the acquired pose graph.

[0047] In some embodiments, the map evaluation device 120 can determine, for each unit area among multiple unit areas, at least one posture associated with the unit area from the posture set, and then determine the relative height value of the unit area based on the height mean of the unit area in the projection map 215 and the corresponding height value of at least one posture.

[0048] At block 430 , the map evaluation device 120 may generate a grayscale image by mapping the plurality of relative height values ​​of the plurality of unit areas into grayscale values, respectively.

[0049] At block 440 , the map evaluation device 120 may generate a gradient map of the grayscale image by performing edge detection on the grayscale image, the gradient map indicating corresponding gradients of the plurality of unit areas.

[0050] At block 450 , the map evaluation device 120 may evaluate the smoothness of the flat area 225 based on the corresponding gradient of the at least one unit area.

[0051] In some embodiments, for each unit area of ​​the at least one unit area, if it is determined that the gradient of the unit area exceeds a first threshold, the map evaluation device 120 may determine the unit area as a smoothness outlier within the flat area 225 .

[0052] Additionally or alternatively, in some embodiments, if the map evaluation device 120 determines that the distance between the smoothness anomaly point of the flat area 225 and the boundary of the flat area 225 is less than a second threshold, the map evaluation device 120 may determine the smoothness anomaly point as a smoothness alarm point within the flat area 225. In some embodiments, if the map evaluation device 120 determines that the distance between the smoothness anomaly point of the flat area 225 and the boundary of the flat area 225 exceeds the second threshold, the map evaluation device 120 may determine the smoothness anomaly point as a smoothness error point within the flat area 225. In some embodiments, the map evaluation device 120 may generate statistical information related to the smoothness alarm points and / or the smoothness error points within the flat area 225.

[0053] The embodiments of the present disclosure also provide corresponding devices for implementing the above-mentioned methods 300 and / or 400. Figure 5 A block diagram of an example apparatus 500 for assessing map quality according to an embodiment of the present disclosure is shown.

[0054] like Figure 5 As shown, the apparatus 500 includes an identification module 510 configured to identify a flat area in the point cloud map based on semantic information of the point cloud map. The apparatus 500 also includes an evaluation module 520 configured to evaluate the quality of the point cloud map by evaluating at least one attribute of the flat area.

[0055] In some embodiments, the evaluation module 520 includes at least one of: a smoothness evaluation module configured to evaluate the smoothness of the flat area; and a thickness evaluation module configured to evaluate the thickness of the flat area.

[0056] In some embodiments, the identification module 510 includes: a projection module, configured to obtain a projection map corresponding to the point cloud map by performing a two-dimensional grid projection on the point cloud map, wherein the projection map is divided into multiple unit areas, each unit area having a height mean and a height variance determined based on the height value of the original coordinate point corresponding to the unit area in the point cloud map; and a flat area identification module, configured to identify a flat area in the projection map based on semantic information, wherein the flat area includes at least one unit area among the multiple unit areas.

[0057] In some embodiments, the smoothness evaluation module includes: an acquisition unit, configured to acquire a pose graph corresponding to a point cloud map, the pose graph including a pose set of an acquisition device when acquiring the point cloud map; a first determination unit, configured to determine multiple relative height values ​​of multiple unit areas based on corresponding height values ​​of the pose set and corresponding height averages of multiple unit areas in the projection map; a second determination unit, configured to determine corresponding gradients of multiple unit areas based on multiple relative height values; and an evaluation unit, configured to evaluate the smoothness of a flat area based on the gradient of at least one unit area.

[0058] In some embodiments, the second determination unit is configured to: generate a grayscale image by mapping multiple relative height values ​​into grayscale values ​​respectively; and generate a gradient map of the grayscale image by performing edge detection on the grayscale image, wherein the gradient map indicates corresponding gradients of multiple unit areas.

[0059] In some embodiments, the evaluation unit is configured to: for each unit area in the at least one unit area, if it is determined that the gradient of the unit area exceeds a first threshold, determine the unit area as a smoothness abnormal point in the flat area.

[0060] In some embodiments, the evaluation unit is further configured to: if it is determined that the distance between the smoothness anomaly point in the flat area and the boundary of the flat area is less than a second threshold, determine the smoothness anomaly point as a smoothness alarm point in the flat area; if it is determined that the distance between the smoothness anomaly point in the flat area and the boundary of the flat area exceeds the second threshold, determine the smoothness anomaly point as a smoothness error point in the flat area; and generate statistical information related to the smoothness alarm point and / or smoothness error point in the flat area.

[0061] In some embodiments, the thickness assessment module includes: a generating unit configured to generate a statistical result based on a corresponding height variance of at least one unit area; and a third determining unit configured to determine whether there is an abnormality in the thickness of the flat area based on the statistical result.

[0062] The modules and units included in the device 500 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units can be implemented using software and / or firmware, such as machine executable instructions stored on a storage medium. In addition to or as an alternative to machine executable instructions, some or all of the modules and units in the device 500 can be implemented at least in part by one or more hardware logic components. As an example and not limitation, exemplary types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0063] Figure 6 6 is a block diagram of an example electronic device 600 capable of implementing various embodiments of the present disclosure. Figure 1 The map evaluation device 120 shown may be implemented by the device 600. Figure 6 As shown, the device 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 602 or computer program instructions loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0064] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0065] The various processes and processing described above, such as methods 300 and / or 400, may be performed by the processing unit 601. For example, in some embodiments, the methods 300 and / or 400 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the CPU 601, one or more actions of the methods 300 and / or 400 described above may be performed.

[0066] The present disclosure may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0067] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of computer readable storage medium include: portable computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination of the above. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0068] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0069] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0070] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0071] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0072] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0073] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0074] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for evaluating map quality, comprising: Based on semantic information of the point cloud map, identifying a flat area in the point cloud map, wherein the flat area includes at least one unit area; and evaluating the quality of the point cloud map by evaluating at least one property of the flat area, wherein evaluating at least one property of the flat area includes evaluating the smoothness of the flat area; The evaluating the smoothness of the flat area comprises: for each unit area of ​​the at least one unit area, if it is determined that the gradient of the unit area exceeds a first threshold, determining the unit area as a smoothness abnormal point in the flat area; The method also includes: if it is determined that the distance between the smoothness abnormal point in the flat area and the boundary of the flat area is less than a second threshold, determining the smoothness abnormal point as a smoothness alarm point in the flat area; if it is determined that the distance between the smoothness abnormal point in the flat area and the boundary of the flat area is greater than the second threshold, determining the smoothness abnormal point as a smoothness error point in the flat area; and generating statistical information related to the smoothness alarm point and / or the smoothness error point in the flat area.

2. The method of claim 1, wherein identifying a flat area in the point cloud map comprises: Obtaining a projection map corresponding to the point cloud map by performing a two-dimensional grid projection on the point cloud map, wherein the projection map is divided into a plurality of unit areas, each unit area having a height mean and a height variance determined based on a height value of an original coordinate point corresponding to the unit area in the point cloud map; as well as Based on the semantic information, the flat area in the projection map is identified, wherein the flat area includes at least one unit area among the plurality of unit areas.

3. The method of claim 2, wherein evaluating the smoothness of the flat area comprises: Acquire a pose graph corresponding to the point cloud map, wherein the pose graph includes a pose set of a collection device when collecting the point cloud map; Determine a plurality of relative height values ​​of the plurality of unit areas based on the corresponding height values ​​of the pose set and the corresponding height averages of the plurality of unit areas in the projection map; Based on the plurality of relative height values, determining corresponding gradients of the plurality of unit areas; as well as The smoothness of the flat area is evaluated based on the gradient of the at least one unit area.

4. The method according to claim 3, wherein determining the respective gradients of the plurality of unit areas comprises: Generate a grayscale image by mapping the plurality of relative height values ​​into grayscale values ​​respectively; as well as A gradient map of the grayscale image is generated by performing edge detection on the grayscale image, wherein the gradient map indicates corresponding gradients of the plurality of unit regions.

5. The method of claim 2, wherein the evaluating at least one property of the planar region comprises evaluating a thickness of the planar region, the method further comprising: generating statistical results based on the corresponding height variance of the at least one unit area; as well as Based on the statistical results, it is determined whether there is an abnormality in the thickness of the flat area.

6. A device for evaluating map quality, comprising: A recognition module, configured to recognize a flat area in the point cloud map based on semantic information of the point cloud map, wherein the flat area includes at least one unit area; as well as An evaluation module configured to evaluate the quality of the point cloud map by evaluating at least one attribute of the flat area, wherein the evaluation module comprises: a smoothness evaluation module configured to evaluate the smoothness of the flat area; For each unit area in the at least one unit area, if it is determined that the gradient of the unit area exceeds a first threshold, the unit area is determined as a smoothness abnormal point in the flat area; If it is determined that the distance between the smoothness abnormal point in the flat area and the boundary of the flat area is less than a second threshold, determining the smoothness abnormal point as a smoothness alarm point in the flat area; If it is determined that the distance between the smoothness abnormal point in the flat area and the boundary of the flat area is greater than the second threshold, determining the smoothness abnormal point as a smoothness error point in the flat area; and Statistical information related to smoothness alarm points and / or smoothness error points within the flat region is generated.

7. The apparatus according to claim 6, wherein the identification module comprises: a projection module configured to obtain a projection map corresponding to the point cloud map by performing a two-dimensional grid projection on the point cloud map, wherein the projection map is divided into a plurality of unit areas, each unit area having a height mean and a height variance determined based on a height value of an original coordinate point corresponding to the unit area in the point cloud map; as well as The flat area recognition module is configured to recognize the flat area in the projection map based on the semantic information, wherein the flat area includes at least one unit area among the plurality of unit areas.

8. The apparatus according to claim 7, wherein the smoothness evaluation module comprises: An acquisition unit is configured to acquire a pose graph corresponding to the point cloud map, wherein the pose graph includes a pose set of a collection device when collecting the point cloud map; A first determining unit is configured to determine a plurality of relative height values ​​of the plurality of unit areas based on corresponding height values ​​of the pose set and corresponding height averages of the plurality of unit areas in the projection map; A second determining unit is configured to determine corresponding gradients of the plurality of unit areas based on the plurality of relative height values; as well as An evaluation unit is configured to evaluate the smoothness of the flat area based on the gradient of the at least one unit area.

9. The apparatus according to claim 8, wherein the second determining unit is configured to: Generate a grayscale image by mapping the plurality of relative height values ​​into grayscale values ​​respectively; and A gradient map of the grayscale image is generated by performing edge detection on the grayscale image, wherein the gradient map indicates corresponding gradients of the plurality of unit regions.

10. The apparatus according to claim 7, wherein the evaluation module comprises a thickness evaluation module configured to evaluate the thickness of the flat area, the thickness evaluation module comprising: A generating unit configured to generate a statistical result based on the corresponding height variance of the at least one unit area; as well as A third determining unit is configured to determine whether there is an abnormality in the thickness of the flat area based on the statistical result.

11. An electronic device, comprising: one or more processors; as well as A memory for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the method according to any one of claims 1 to 5.

12. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method according to any one of claims 1 to 5.

Images