Road surface material identification method, device, equipment and medium

Abstract

The application relates to the technical field of road surface recognition, and particularly provides a road surface material recognition method, device, equipment and medium, aiming to solve the problem that in the prior art, when a fixed reflection intensity is used to judge road surface material, the road surface point cloud data is prone to holes, misjudgment and boundary deviation when dry-wet changes, material switching and special areas exist, and the road material recognition error rate is high. To this end, the road surface material recognition method comprises the following steps: acquiring point cloud data of a road surface to be recognized; wherein the point cloud data comprises reflection intensity and three-dimensional coordinates; inputting the reflection intensity into a preset Gaussian mixture model to obtain a material peak value of the road surface to be recognized; and determining material information of the road surface to be recognized according to the material peak value of the road surface to be recognized.

Description

Technical Field

[0001] This application relates to the field of road surface recognition technology, specifically providing a road surface material recognition method, device, equipment, and medium. Background Technology

[0002] Currently, with the rapid development of intelligent transportation and autonomous driving technologies, the requirements for the digital accuracy of road infrastructure are increasing. Among these requirements, road surface material is a core element of road attribute information, and accurate identification of road surface material is of great significance for autonomous driving road perception, map building, and 3D annotation.

[0003] In existing technologies, different road surface materials exhibit significant differences in reflectance intensity, meaning that the point cloud intensity distribution varies across different materials. Therefore, a fixed reflectance intensity threshold can be used to identify the road surface material. However, reflectance intensity is also affected by factors such as temperature, humidity, rainfall, incident angle, distance, and sensor status. Using a fixed reflectance intensity threshold can lead to issues like voids, misjudgments, and boundary shifts in road surface point cloud data when there are changes in humidity, material switching, or the presence of special areas, resulting in a high error rate in road material identification.

[0004] Accordingly, a new method for identifying road surface materials is needed in this field to address the aforementioned problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects, this application is made to provide a solution or at least a partial solution to the problem in the prior art that when using a fixed reflection intensity to determine the road surface material, the road surface point cloud data is prone to holes, misjudgments and boundary offsets when there are changes in dryness and wetness, material switching and the existence of special areas, which leads to a high error rate in road material identification.

[0006] Firstly, this application provides a method for identifying road surface materials, including:

[0007] Acquire point cloud data of the road surface to be identified; wherein, the point cloud data includes reflection intensity and three-dimensional coordinates;

[0008] The reflection intensity is input into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified;

[0009] The material information of the road surface to be identified is determined based on the material peak value of the road surface to be identified.

[0010] In one technical solution of the above-mentioned road surface material identification method, the step of acquiring point cloud data of the road surface to be identified includes:

[0011] The original point cloud data, environmental data, and lidar parameters of the road surface to be identified are acquired; wherein, the original point cloud data includes the original reflection intensity and three-dimensional coordinates;

[0012] Based on the environmental data and the lidar parameters, the original reflection intensity in the original point cloud data is compensated to obtain the point cloud data of the road surface to be identified.

[0013] In one technical solution of the above-mentioned road surface material identification method, the environmental data includes humidity and temperature, and the lidar parameters include incident angle and distance. The step of compensating for the original reflection intensity in the original point cloud data based on the environmental data and the lidar parameters to obtain the point cloud data of the road surface to be identified includes:

[0014] Obtain the preliminary material type of the road surface to be identified;

[0015] The original reflection intensity is compensated according to the humidity compensation coefficient corresponding to the humidity to obtain the humidity-compensated reflection intensity;

[0016] Based on the degree of influence of the temperature on the initial material type, the humidity-compensated reflection intensity is compensated to obtain the temperature-compensated reflection intensity;

[0017] Based on the incident angle, the temperature-compensated reflection intensity is geometrically corrected to obtain the angle-compensated reflection intensity.

[0018] Based on the distance, the angle-compensated reflection intensity is attenuated to obtain the reflection intensity;

[0019] The original reflection intensity in the original point cloud data is replaced by the reflection intensity to obtain the point cloud data of the road surface to be identified.

[0020] In one technical solution of the above-mentioned road surface material identification method, the step of inputting the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified includes:

[0021] According to a preset clustering algorithm, the reflection intensity is clustered to obtain the cluster centers and variances of various reflection intensities;

[0022] Based on the cluster centers and variances of the various reflection intensities, the parameters of the preset Gaussian mixture model are updated to obtain the material peak value of the road surface to be identified.

[0023] In one technical solution of the above-mentioned road surface material identification method, determining the material information of the road surface to be identified based on the material peak value of the road surface to be identified includes:

[0024] Determine multi-level adaptive thresholds based on the valley value between two adjacent material peaks;

[0025] The reflection intensity of each point in the point cloud data is matched with the multi-level adaptive threshold to obtain the material information of each point in the point cloud data.

[0026] In one technical solution of the above-mentioned road surface material identification method, the method further includes:

[0027] Obtain the gradient data of the point cloud data;

[0028] The material boundaries are determined based on the gradient data of the point cloud data;

[0029] Based on the material boundaries, different material regions in the point cloud data are determined;

[0030] By merging isolated regions within the same material region with other regions within the same material region, a connected region is obtained;

[0031] The material of each point in the connected region is processed in the same way to obtain segmented point cloud data;

[0032] Based on the reflection intensity of the connected regions in the segmented point cloud data, the dangerous areas in the road surface to be identified are obtained.

[0033] In one technical solution of the above-mentioned road surface material identification method, obtaining the dangerous area in the road surface to be identified based on the reflection intensity of the connected regions in the segmented point cloud data includes:

[0034] If the reflection intensity of a connected region in the segmented point cloud data is less than a first preset reflection intensity, then the connected region is a water accumulation region.

[0035] If the reflection intensity of a connected region in the segmented point cloud data is greater than a second preset reflection intensity, then the connected region is an icing region.

[0036] If the reflection intensity of a connected region in the segmented point cloud data is determined to be greater than the first preset reflection intensity and less than the second preset reflection intensity, then the connected region is a slippery risk area.

[0037] Secondly, this application provides a road surface material identification device, comprising:

[0038] An acquisition module is used to acquire point cloud data of the road surface to be identified; wherein, the point cloud data includes reflection intensity and three-dimensional coordinates;

[0039] The processing module is used to input the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified;

[0040] The determination module is used to determine the material information of the road surface to be identified based on the material peak value of the road surface to be identified.

[0041] Thirdly, this application provides a road surface material identification device, including at least one processor and a memory; wherein,

[0042] The memory stores computer-executed instructions;

[0043] The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method described in any one of the first aspects.

[0044] Fourthly, this application provides a computer-readable storage medium storing a plurality of program codes adapted to be loaded and run by a processor to perform the method described in any one of the first aspects.

[0045] This application provides a method, apparatus, device, and medium for identifying road surface materials. The method specifically involves: acquiring point cloud data of the road surface to be identified; wherein the point cloud data includes reflection intensity and three-dimensional coordinates; inputting the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified; and determining the material information of the road surface to be identified based on the material peak value, thereby improving the accuracy of road surface material identification. Attached Figure Description

[0046] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings are used to denote similar components, wherein:

[0047] Figure 1 This is a flowchart illustrating a method for identifying road surface materials according to an embodiment of this application.

[0048] Figure 2 This is a flowchart illustrating a second embodiment of a road surface material identification method provided in this application.

[0049] Figure 3 This is a flowchart illustrating a third embodiment of a road surface material identification method provided in this application.

[0050] Figure 4 This is a flowchart illustrating a fourth embodiment of a road surface material identification method provided in this application.

[0051] Figure 5 This is a flowchart illustrating a fifth embodiment of a road surface material identification method provided in this application.

[0052] Figure 6 This is a flowchart illustrating a sixth embodiment of a road surface material identification method provided in this application.

[0053] Figure 7 This is a flowchart illustrating Embodiment Seven of a road surface material identification method provided in this application.

[0054] Figure 8 This is a schematic diagram of the structure of a road surface material identification device provided in an embodiment of this application;

[0055] Figure 9 This is a schematic diagram of the structure of a road surface material identification device provided in an embodiment of this application.

[0056] List of reference numerals in the attached diagram:

[0057] 11: Acquisition module; 12: Processing module; 13: Determination module; 21: Processor; 22: Memory. Detailed Implementation

[0058] Some embodiments of this application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of this application and are not intended to limit the scope of protection of this application.

[0059] In the description of this application, "module" and "processor" can include hardware, software, or a combination of both. A module can include hardware circuitry, various suitable sensors, communication ports, memory, and can also include software components, such as program code, or a combination of software and hardware. A processor can be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and / or signal processing capabilities. The processor can be implemented in software, in hardware, or a combination of both. Non-transitory computer-readable storage media includes any suitable medium capable of storing program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The terms "at least one A or B" or "at least one of A and B" have a similar meaning to "A and / or B" and can include only A, only B, or A and B. The singular terms "a" or "this" can also include plural forms.

[0060] In existing technologies, point cloud data of roads is collected using LiDAR, and then the reflection intensity in the point cloud data is compared with a preset fixed reflection intensity threshold to determine the road surface material. However, reflection intensity is affected by factors such as temperature, humidity, rainfall, incident angle, distance, and sensor status. If a fixed reflection intensity threshold is used, the road surface point cloud data is prone to holes, misjudgments, and boundary offsets when there are changes in wet and dry conditions, material switching, or the presence of special areas, leading to a high error rate in road material identification.

[0061] Based on this, in order to solve the above-mentioned technical problems, the technical concept of this application is to provide a new method for road surface material identification in order to improve the accuracy of road surface material identification.

[0062] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0063] Figure 1 This is a flowchart illustrating an embodiment of a road surface material identification method provided in this application. Figure 1 As shown, specifically, the method includes:

[0064] Step S101: Obtain point cloud data of the road surface to be identified.

[0065] In this embodiment, the point cloud data includes reflection intensity and three-dimensional coordinates. The reflection intensity is the energy signal of the laser echo, reflecting the surface material, roughness, and incident angle characteristics of the road. The three-dimensional coordinates are the calculated Cartesian space position, reflecting the object's geometric structure.

[0066] In this embodiment, point cloud data can be collected using lidar, as well as data such as vehicle pose, location, and collection time.

[0067] Step S102: Input the reflection intensity into the preset Gaussian mixture model to obtain the material peak value of the road surface to be identified.

[0068] In this embodiment, the reflection intensity of the same road surface material varies under different environments. After inputting the current road surface reflection intensity into a preset Gaussian mixture model, the model classifies the reflection intensity to obtain the material peak values ​​corresponding to various road surface materials. Specifically, asphalt has low reflection intensity with a material peak value ranging from 0.1 to 0.2; cement has medium reflection intensity with a material peak value ranging from 0.4 to 0.6; road markings have high reflection intensity with a material peak value ranging from 0.7 to 0.9; standing water has extremely low reflection intensity with a material peak value ranging from 0.02 to 0.05; and ice has abnormally high reflection intensity with a material peak value greater than 1.

[0069] Step S103: Determine the material information of the road surface to be identified based on the material peak value.

[0070] In this embodiment, a reflection intensity threshold is determined based on the peak value of the material of the road surface to be identified, and the material of the road surface to be identified is classified according to the reflection intensity threshold.

[0071] In this embodiment, point cloud data of the road surface to be identified is acquired; wherein, the point cloud data includes reflection intensity and three-dimensional coordinates; the reflection intensity is input into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified; based on the material peak value of the road surface to be identified, the material information of the road surface to be identified is determined. Compared with the prior art, when using a fixed reflection intensity to determine the road surface material, the road point cloud data is prone to voids, misjudgments, and boundary offsets when there are changes in dryness and wetness, material switching, and the existence of special areas, resulting in a high road material identification error rate, this application inputs the reflection intensity of the road surface to be identified into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified, and uses the material peak value of the road surface to be identified to determine the material information of the road surface to be identified. For all roads surface to be identified, the Gaussian mixture model is used for material analysis, and the road surface material is determined based on the analysis results, thereby improving the accuracy of road surface material identification.

[0072] Figure 2 This is a schematic flowchart of a second embodiment of a road surface material identification method provided in this application. Figure 2 As shown, the specific implementation of step S101 includes:

[0073] Step S201: Obtain the original point cloud data, environmental data, and lidar parameters of the road surface to be identified.

[0074] In this embodiment, the raw point cloud data includes the raw reflection intensity and three-dimensional coordinates.

[0075] Step S202: Based on environmental data and lidar parameters, compensate for the original reflection intensity in the original point cloud data to obtain the point cloud data of the road surface to be identified.

[0076] In this embodiment, the original reflection intensity is affected by the environment and lidar parameters, which may reduce the accuracy of material identification of the road surface to be identified. Therefore, it is necessary to compensate and correct the original reflection intensity based on the temperature and humidity in the environmental data and the incident angle and distance in the lidar parameters to obtain accurate point cloud data of the road surface to be identified.

[0077] In this embodiment, the original point cloud data of the road surface to be identified, environmental data, and lidar parameters are acquired. The original point cloud data includes the original reflection intensity and three-dimensional coordinates. Based on the environmental data and lidar parameters, the original reflection intensity in the original point cloud data is compensated to obtain the point cloud data of the road surface to be identified. By compensating for the original reflection intensity in the original point cloud data, the adverse effects of the external environment and lidar parameters on the reflection intensity are eliminated, and the accurate reflection intensity of the road surface material can be obtained. Based on this accurate reflection intensity, the identification accuracy of the road surface material can be improved.

[0078] Figure 3 This is a flowchart illustrating a third embodiment of a road surface material identification method provided in this application. Figure 3 As shown, specifically, environmental data includes humidity and temperature, and lidar parameters include incident angle and distance. Therefore, the implementation of step S202 includes:

[0079] Step S301: Obtain the preliminary material type of the road surface to be identified.

[0080] In this embodiment, the possible preliminary material type of the road surface to be identified can be obtained from a priori road surface material map or historical database based on the location information of the road surface to be identified.

[0081] Step S302: Compensate the original reflection intensity according to the humidity compensation coefficient corresponding to the humidity to obtain the humidity-compensated reflection intensity.

[0082] In this embodiment, a humidity compensation coefficient corresponding to a preset humidity range is used. Based on the humidity in the current environmental data, the corresponding humidity range is determined, and the humidity compensation coefficient corresponding to that humidity range is also determined. Since increased humidity causes laser echo energy attenuation, reducing the original reflection intensity, it is necessary to compensate for the original reflection intensity according to the humidity compensation coefficient to obtain a more accurate reflection intensity.

[0083] In this embodiment, the humidity-compensated reflection intensity is obtained by multiplying the humidity compensation coefficient by the original reflection intensity.

[0084] Step S303: Based on the influence of temperature on the initial material type, compensate for the humidity-compensated reflection intensity to obtain the temperature-compensated reflection intensity.

[0085] In this embodiment, temperature affects the hardware parameters of the lidar, as well as the physical reflectivity of the atmosphere and the road surface. Therefore, it is necessary to compensate for the humidity-compensated reflection intensity based on the temperature.

[0086] In this embodiment, an influence degree corresponding to a temperature range is preset. Based on the temperature in the environmental data, the temperature range corresponding to the temperature is determined. Then, based on the temperature range, the influence degree corresponding to the temperature range is determined. The temperature-compensated reflection intensity can be obtained by multiplying the influence degree by the humidity-compensated reflection intensity.

[0087] Step S304: Based on the incident angle, perform geometric correction on the temperature-compensated reflection intensity to obtain the angle-compensated reflection intensity.

[0088] In this embodiment, since the incident angle of the laser irradiation on the road surface to be identified is different, resulting in different echo energy differences, it is necessary to perform geometric correction on the temperature compensation reflection intensity according to the incident angle.

[0089] In this embodiment, specifically, the angle-compensated reflection intensity can be obtained by multiplying the incident angle compensation function and the temperature-compensated reflection intensity.

[0090] Step S305: Based on the distance, perform attenuation compensation on the angle-compensated reflection intensity to obtain the reflection intensity.

[0091] In this embodiment, a standard Lambertian reflective whiteboard with known reflectivity is first used to collect multiple sets of echo intensity data at different calibrated distances. By fitting these multiple sets of echo intensity data, an attenuation baseline curve between distance and original intensity is obtained. The least squares method is then used to fit this attenuation baseline curve, resulting in a fitting formula. Substituting the distance into this fitting formula yields the reflection intensity.

[0092] Step S306: Replace the original reflection intensity in the original point cloud data with the reflection intensity to obtain the point cloud data of the road surface to be identified.

[0093] In this embodiment, the preliminary material type of the road surface to be identified is obtained; the original reflection intensity is compensated according to the humidity compensation coefficient corresponding to the humidity to obtain the humidity-compensated reflection intensity; the humidity-compensated reflection intensity is compensated according to the influence of temperature on the preliminary material type to obtain the temperature-compensated reflection intensity; the temperature-compensated reflection intensity is geometrically corrected according to the incident angle to obtain the angle-compensated reflection intensity; the angle-compensated reflection intensity is attenuated according to the distance to obtain the reflection intensity; the reflection intensity is replaced with the original reflection intensity in the original point cloud data to obtain the point cloud data of the road surface to be identified. By compensating the original reflection intensity through temperature, humidity, incident angle, and distance, the error in reflection intensity caused by the external environment and lidar parameters can be corrected, thereby improving the accuracy of the reflection intensity.

[0094] Figure 4 This is a flowchart illustrating Embodiment 4 of a road surface material identification method provided in this application. Figure 4 As shown, the specific implementation of step S102 includes:

[0095] Step S401: Cluster the reflection intensity according to the preset clustering algorithm to obtain the cluster centers and variances of various reflection intensities.

[0096] In this embodiment, the reflection intensity is input to a preset clustering algorithm. The preset clustering algorithm clusters the reflection intensity to obtain the cluster center and variance of each reflection intensity. The initialization model is then constructed using the cluster center and variance.

[0097] Step S402: Update the parameters of the preset Gaussian mixture model based on the cluster centers and variances of various reflection intensities to obtain the material peak value of the road surface to be identified.

[0098] In the expected step, the posterior probability of each reflection intensity belonging to the k-th Gaussian component is calculated to obtain the responsibility degree. Then, the mean, variance, and weights of the initial model are updated using this responsibility degree. This process is repeated until the parameter change is less than a preset threshold or the maximum number of iterations is reached, at which point the iteration stops, and the material peak value of the road surface to be identified is finally output.

[0099] In this embodiment, the reflection intensity is clustered according to a preset clustering algorithm to obtain the cluster centers and variances of various reflection intensities. Based on the cluster centers and variances of various reflection intensities, the parameters of the preset Gaussian mixture model are updated to obtain the material peak value of the road surface to be identified. By utilizing the model, material peak values ​​adapted to different road surface materials can be generated, thereby reducing the workload of manually adjusting the reflection intensity threshold and improving the efficiency of road surface material identification.

[0100] Figure 5 This is a flowchart illustrating Embodiment 5 of a road surface material identification method provided in this application. Figure 5 As shown, the specific implementation of step S103 includes:

[0101] Step S501: Determine the multi-level adaptive threshold based on the valley value between two adjacent material peaks.

[0102] In this embodiment, multiple threshold intervals are determined based on the valley value between two adjacent material peaks, and the set of intermediate values ​​of these threshold intervals is a multi-level adaptive threshold.

[0103] In this embodiment, the multi-level adaptive threshold includes a primary threshold and a secondary threshold. The primary threshold can distinguish between road surfaces and non-road surfaces.

[0104] Step S502: Match the reflection intensity of each point in the point cloud data with a multi-level adaptive threshold to obtain the material information of each point in the point cloud data.

[0105] Specifically, the reflection intensity of each point in the point cloud data is matched with a multi-level adaptive threshold. If the reflection intensity is greater than the maximum value or less than the minimum value among the multi-level adaptive thresholds, the point is classified as non-road surface; otherwise, it is classified as road surface. Then, the points classified as road surface are used to accurately distinguish the road surface material through a secondary threshold.

[0106] In this embodiment, the material information includes the material label and confidence level.

[0107] In this embodiment, a multi-level adaptive threshold is determined based on the valley value between two adjacent material peaks; the reflection intensity of each point in the point cloud data is matched with the multi-level adaptive threshold to obtain the material information of each point in the point cloud data. By determining the multi-level adaptive threshold and obtaining the material information corresponding to each point cloud data according to the multi-level adaptive threshold, the accuracy of road surface material identification is improved.

[0108] Figure 6 This is a flowchart illustrating Embodiment Six of a road surface material identification method provided in this application. Figure 6 As shown, specifically, after step S103, the method further includes:

[0109] Step S601: Obtain gradient data from the point cloud data.

[0110] In this embodiment, the spatial derivative of the reflection intensity is first calculated on the point cloud data to obtain the gradient data of the point cloud data.

[0111] Step S602: Determine the material boundary based on the gradient data of the point cloud data.

[0112] In this embodiment, by comparing the gradient data of the point cloud data with a preset gradient threshold, the part of the point cloud data where the gradient changes is connected to obtain the material boundary.

[0113] Step S603: Determine the different material regions in the point cloud data based on the material boundaries.

[0114] The multiple point cloud datasets enclosed by material boundaries are analyzed separately to determine the different material regions within the point cloud datasets.

[0115] Step S604: Merge isolated regions within the same material region with other regions within the same material region to obtain connected regions.

[0116] In this embodiment, road surfaces of the same material should be continuous areas. Therefore, isolated areas of the same material are merged to obtain connected areas.

[0117] Step S605: Process the material of each point in the connected region in the same way to obtain segmented point cloud data.

[0118] In this embodiment, point cloud data containing individual materials of other materials are considered to have the same material within the same plane. Therefore, the material labels of the point cloud data containing the other materials are set as the material labels of the majority of the point cloud data.

[0119] Step S606: Based on the reflection intensity of the connected regions in the segmented point cloud data, obtain the dangerous areas in the road surface to be identified.

[0120] In this embodiment, the reflection intensity of connected regions in the segmented point cloud data is compared with a preset danger zone threshold to obtain the danger zone in the road surface to be identified.

[0121] In this embodiment, gradient data of point cloud data is acquired; material boundaries are determined based on the gradient data of point cloud data; different material regions in point cloud data are determined based on the material boundaries; isolated regions in the same material region are merged with other regions in the same material region to obtain connected regions; the material of each point in the connected region is processed in the same way to obtain segmented point cloud data; dangerous areas in the road surface to be identified are obtained based on the reflection intensity of the connected regions in the segmented point cloud data, thereby reducing isolated misjudgments and road surface voids.

[0122] Figure 7 This is a flowchart illustrating Embodiment Seven of a road surface material identification method provided in this application. Figure 7 As shown, the specific implementation of step S606 includes:

[0123] Step S701: If the reflection intensity of a connected region in the segmented point cloud data is less than the first preset reflection intensity, then the connected region is a water accumulation region.

[0124] In this embodiment, for example, if the first preset reflection intensity is 0.05, then in the connected regions of the segmented point cloud data, if more than 60% of the points have a reflection intensity of less than 0.05, then the connected region is a waterlogged region.

[0125] Step S702: If the reflection intensity of a connected region in the segmented point cloud data is greater than the second preset reflection intensity, then the connected region is an icing region.

[0126] In this embodiment, for example, if the second preset reflection intensity is 1, then in the connected regions of the segmented point cloud data, if more than 60% of the points have a reflection intensity greater than 1, then the connected region is an icing region.

[0127] Step S703: If the reflection intensity of a connected region in the segmented point cloud data is greater than the first preset reflection intensity and less than the second preset reflection intensity, then the connected region is a slippery risk area.

[0128] In this embodiment, for example, in a connected region of segmented point cloud data, if more than 60% of the points have a reflection intensity greater than 0.05 and less than 1, then the connected region is a slippery risk region.

[0129] In this embodiment, if the reflection intensity of a connected region in the segmented point cloud data is less than a first preset reflection intensity, then the connected region is a water accumulation area; if the reflection intensity of a connected region in the segmented point cloud data is greater than a second preset reflection intensity, then the connected region is an icy area; if the reflection intensity of a connected region in the segmented point cloud data is greater than the first preset reflection intensity and less than the second preset reflection intensity, then the connected region is a slippery risk area. This distinguishes special areas and provides a basis for safety risk warnings and subsequent planning and control.

[0130] It should be noted that although the steps in the above embodiments are described in a specific order, those skilled in the art will understand that in order to achieve the effect of this application, different steps do not necessarily have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders, and these variations are all within the scope of protection of this application.

[0131] Furthermore, this application also provides a road surface material identification device.

[0132] Figure 8 This is a schematic diagram of a road surface material identification device provided in an embodiment of this application. Figure 8 As shown, the road surface material identification device in this embodiment mainly includes an acquisition module 11, a processing module 12, and a determination module 13. In some embodiments, one or more of the acquisition module 11, processing module 12, and determination module 13 can be combined into a single module. In some embodiments, the acquisition module 11 can be configured to acquire point cloud data of the road surface to be identified; wherein the point cloud data includes reflection intensity and three-dimensional coordinates. The processing module 12 can be configured to input the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified. The determination module 13 can be configured to determine the material information of the road surface to be identified based on the material peak value of the road surface to be identified.

[0133] The aforementioned road surface material identification device is used for performing Figures 1 to 7 The road surface material identification method embodiments shown are similar in technical principle, the technical problems solved and the technical effects produced. Those skilled in the art can clearly understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the road surface material identification device can be referred to the content described in the embodiments of the road surface material identification method, which will not be repeated here.

[0134] Those skilled in the art will understand that all or part of the processes in the method of the above-described embodiment can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium can include any entity or device capable of carrying the computer program code, a medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory, a random access memory, an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0135] Furthermore, this application also provides a road surface material identification device.

[0136] Figure 9 This is a schematic diagram of the structure of a road surface material identification device provided in an embodiment of this application. Figure 9 As shown, the road surface material identification device includes a processor 21 and a memory 22. The memory 22 can be configured to store a program for executing the road surface material identification method of the above-described method embodiments. The processor 21 can be configured to execute the program stored in the memory, which includes, but is not limited to, a program for executing the road surface material identification method of the above-described method embodiments. For ease of explanation, only the parts related to the embodiments of this application are shown. For specific technical details not disclosed, please refer to the method section of the embodiments of this application. The road surface material identification device can be a control device device comprising various electronic devices.

[0137] Furthermore, this application also provides a computer-readable storage medium. In one embodiment of the computer-readable storage medium according to this application, the computer-readable storage medium can be configured to store a program that performs the road surface material identification method of the above-described method embodiments. This program can be loaded and run by a processor to implement the above-described road surface material identification method. For ease of explanation, only the parts related to the embodiments of this application are shown; for specific technical details not disclosed, please refer to the method section of the embodiments of this application. The computer-readable storage medium can be a storage device device including various electronic devices. Optionally, in the embodiments of this application, the computer-readable storage medium is a non-transitory computer-readable storage medium.

[0138] Furthermore, it should be understood that since the various modules are only provided to illustrate the functional units of the device described in this application, the physical devices corresponding to these modules may be the processor itself, or a part of the processor's software, hardware, or a combination of both. Therefore, the number of modules shown in the figures is merely illustrative.

[0139] Those skilled in the art will understand that the various modules in the device can be adaptively split or combined. Such splitting or combining of specific modules will not cause the technical solution to deviate from the principles of this application; therefore, the technical solutions after splitting or combining will fall within the protection scope of this application.

[0140] The technical solutions of this application have been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of this application is obviously not limited to these specific embodiments. Without departing from the principles of this application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of this application.

Claims

1. A method for identifying road surface materials, characterized in that, include: Acquire point cloud data of the road surface to be identified; wherein, the point cloud data includes reflection intensity and three-dimensional coordinates; The reflection intensity is input into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified; The material information of the road surface to be identified is determined based on the material peak value of the road surface to be identified.

2. The method according to claim 1, characterized in that, The acquisition of point cloud data of the road surface to be identified includes: The original point cloud data, environmental data, and lidar parameters of the road surface to be identified are acquired; wherein, the original point cloud data includes the original reflection intensity and three-dimensional coordinates; Based on the environmental data and the lidar parameters, the original reflection intensity in the original point cloud data is compensated to obtain the point cloud data of the road surface to be identified.

3. The method according to claim 2, characterized in that, The environmental data includes humidity and temperature, and the lidar parameters include incident angle and distance. The process of compensating for the original reflection intensity in the original point cloud data based on the environmental data and the lidar parameters to obtain point cloud data of the road surface to be identified includes: Obtain the preliminary material type of the road surface to be identified; The original reflection intensity is compensated according to the humidity compensation coefficient corresponding to the humidity to obtain the humidity-compensated reflection intensity; Based on the degree of influence of the temperature on the initial material type, the humidity-compensated reflection intensity is compensated to obtain the temperature-compensated reflection intensity; Based on the incident angle, the temperature-compensated reflection intensity is geometrically corrected to obtain the angle-compensated reflection intensity. Based on the distance, the angle-compensated reflection intensity is attenuated to obtain the reflection intensity; The original reflection intensity in the original point cloud data is replaced by the reflection intensity to obtain the point cloud data of the road surface to be identified.

4. The method according to claim 1, characterized in that, The step of inputting the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified includes: According to a preset clustering algorithm, the reflection intensity is clustered to obtain the cluster centers and variances of various reflection intensities; Based on the cluster centers and variances of the various reflection intensities, the parameters of the preset Gaussian mixture model are updated to obtain the material peak value of the road surface to be identified.

5. The method according to claim 1, characterized in that, The step of determining the material information of the road surface to be identified based on the material peak value includes: Determine multi-level adaptive thresholds based on the valley value between two adjacent material peaks; The reflection intensity of each point in the point cloud data is matched with the multi-level adaptive threshold to obtain the material information of each point in the point cloud data.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the gradient data of the point cloud data; The material boundaries are determined based on the gradient data of the point cloud data; Based on the material boundaries, different material regions in the point cloud data are determined; By merging isolated regions within the same material region with other regions within the same material region, a connected region is obtained; The material of each point in the connected region is processed in the same way to obtain segmented point cloud data; Based on the reflection intensity of the connected regions in the segmented point cloud data, the dangerous areas in the road surface to be identified are obtained.

7. The method according to claim 6, characterized in that, The step of obtaining the hazardous area in the road surface to be identified based on the reflection intensity of the connected regions in the segmented point cloud data includes: If the reflection intensity of a connected region in the segmented point cloud data is less than a first preset reflection intensity, then the connected region is a water accumulation region. If the reflection intensity of a connected region in the segmented point cloud data is greater than a second preset reflection intensity, then the connected region is an icing region. If the reflection intensity of a connected region in the segmented point cloud data is determined to be greater than the first preset reflection intensity and less than the second preset reflection intensity, then the connected region is a slippery risk area.

8. A road surface material identification device, characterized in that, include: An acquisition module is used to acquire point cloud data of the road surface to be identified; wherein, the point cloud data includes reflection intensity and three-dimensional coordinates; The processing module is used to input the reflection intensity into a preset Gaussian mixture model to obtain the material peak value of the road surface to be identified; The determination module is used to determine the material information of the road surface to be identified based on the material peak value of the road surface to be identified.

9. A road surface material identification device, comprising at least one processor and a memory; wherein, The memory stores computer-executed instructions; The at least one processor executes computer execution instructions stored in the memory, causing the at least one processor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by a processor to perform the method of any one of claims 1 to 7.

Images