Lightweight three-dimensional detection method and device suitable for rural roads

Abstract

The present invention provides a lightweight three-dimensional inspection method and device suitable for rural roads. The method comprises: obtaining original three-dimensional contour data of the rural road pavement using a line scanning three-dimensional measurement sensor; extracting target road area data and positions based on the original three-dimensional contour data of the rural road pavement, thereby obtaining three-dimensional modeling data of the target road area; and performing road surface inspection on the target road area to obtain inspection results. The lightweight three-dimensional inspection method suitable for rural roads of the present invention obtains three-dimensional contour data of the road using a line scanning three-dimensional measurement sensor, then determines the actual target road area data and positions based on various characteristics of the road, and then establishes three-dimensional modeling data of the target road area to detect road surface defects, etc. The inspection operation is convenient, the results are accurate, and the method is suitable for large-scale inspection scenarios of rural roads.

Description

Technical Field

[0001] The present invention relates to the field of road detection technology, and in particular to a lightweight three-dimensional detection method and device suitable for rural roads. Background Art

[0002] Rural roads have entered a new stage of development, prioritizing both construction and maintenance. By the end of 2020, the mileage of rural roads reached 4.38 million kilometers, with 3.81 million kilometers assessed. Large-scale, fully automated inspections of rural roads are urgently needed. Road condition data is essential for improving the safety of rural road traffic facilities and making informed decisions about large-scale maintenance.

[0003] To meet the large-scale demand for rural road inspections, existing highway infrastructure condition monitoring technologies are primarily targeted at high-grade expressways. Due to the poor capacity, wide variations in road width, complex terrain, and widely varying road conditions of rural roads, many lack a technically sound, economically feasible, and reliable inspection method. Designing a reliable rural road inspection method is an urgent challenge. Summary of the Invention

[0004] The present invention provides a lightweight three-dimensional detection method and device suitable for rural roads, which are used to solve the defect that rural roads are difficult to detect in the prior art and realize low-cost and high-precision detection of rural roads.

[0005] The present invention provides a lightweight three-dimensional detection method applicable to rural roads, comprising:

[0006] Using a line scanning 3D measurement sensor to obtain original rural highway pavement 3D contour data; the original rural highway pavement 3D contour data is point cloud data;

[0007] Based on the original rural highway pavement 3D profile data, the target pavement area data and location are extracted by utilizing the macroscopic continuity characteristics of the pavement surface profile, the characteristics of the non-pavement area being located on both sides of the road width, the elevation mutation characteristics of the adjacent positions of the pavement area and the non-pavement area, and the continuity characteristics of the measurement posture;

[0008] Extracting measurement posture information based on the road surface target area data, and eliminating the measurement posture influence from the original rural highway road surface three-dimensional contour data to obtain three-dimensional modeling data of the road surface target area;

[0009] Determining a road surface type based on the three-dimensional modeling data of the road surface target area; the road surface type includes at least one of an asphalt road surface, a cement road surface, and a gravel road surface;

[0010] Based on the three-dimensional modeling data of the target road area and the determined road type, a road surface detection is performed on the target road area to obtain a detection result; the road surface detection includes at least one of road damage detection, road surface flatness detection and road surface rutting detection.

[0011] According to the present invention, a lightweight three-dimensional detection method suitable for rural roads is provided. Based on the original three-dimensional contour data of the rural road surface, the method utilizes the macroscopic continuity characteristics of the apparent contour of the road surface, the characteristics of the non-road surface area being located on both sides of the road width, the characteristics of the elevation mutation at adjacent positions of the road surface area and the non-road surface area, and the continuity characteristics of the measurement posture to extract the data and position of the target area of ​​the road surface, including:

[0012] Based on the original rural highway pavement three-dimensional profile data, for all measurement points in any cross section, the elevation difference between two measurement points with a first preset distance between the measurement points is calculated from the inside to the outside;

[0013] If the elevation difference is greater than a first preset threshold, the measuring point located outside is marked as an abnormal mutation point;

[0014] Generate a binary graph based on each abnormal mutation point;

[0015] The binary image is extended using a morphological processing method, and denoising is performed based on the area and length characteristics of the connected regions in the extended binary image to determine the data and position of the target road area.

[0016] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, the method extracts measurement posture information based on the road surface target area data, eliminates the measurement posture from the original rural road road surface three-dimensional contour data, and obtains three-dimensional modeling data of the road surface target area, including:

[0017] Obtaining the difference between the average elevation of the starting cross section and the average elevation of the ending cross section corresponding to any cross section in the road target area data, determining the cross section with the difference greater than a preset threshold as a jump surface, processing the road target area according to all jump surfaces, and obtaining all divided road sections;

[0018] Constructing first time series data based on the average elevation of all cross sections of each divided road section, converting the first time series data into first frequency data, and obtaining the pitch and vibration elevation of the corresponding measurement posture of each cross section based on the first frequency data;

[0019] Constructing second time series data based on the slopes of all cross sections of each divided road section, converting the second time series data into second frequency data, and obtaining a roll angle corresponding to a measured posture of each cross section based on the second frequency data;

[0020] Correcting the road target area data according to the pitch and vibration elevation of the cross-section corresponding to the measured posture and the roll angle of the cross-section corresponding to the measured posture to obtain three-dimensional modeling data of the road target area;

[0021] The starting cross-section is the corresponding cross-section taken forward with the cross-section as the center plane along the data collection sequence and at intervals of the first preset length; the ending cross-section is the corresponding cross-section taken backward with the cross-section as the center plane along the data collection sequence and at intervals of the first preset length.

[0022] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, determining the type of the road surface based on the three-dimensional modeling data of the target road surface area includes:

[0023] Calculating the structural depth of each measuring point in the target road surface area based on the three-dimensional road surface modeling data of the target road surface area;

[0024] Obtaining a road surface texture feature of the target road surface area according to the structural depth of each measurement point in the target road surface area; the road surface texture feature includes uniformity, periodicity, and size data of the road surface texture distribution;

[0025] The type of the road surface is determined based on the road surface texture characteristics of the target road surface area.

[0026] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, the road surface flatness detection includes the following steps:

[0027] Determining longitudinal profile data of a preset position based on the three-dimensional road surface modeling data of the road surface target area;

[0028] When the preset position is located within the road surface target area, calculating an International Roughness Index (IRI) based on the longitudinal profile data of the preset position;

[0029] In a case where the preset position is not located within the road surface target area, the preset position is marked.

[0030] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, the road surface damage detection comprises the following steps:

[0031] Inputting the three-dimensional modeling data of the target road surface area into a defect recognition model to obtain point cloud data of the defect area output by the defect recognition model; the defect recognition model is trained using the point cloud data of the defective road surface as a sample and the point cloud data of the defect area in the defective road surface as a label;

[0032] Determining the location of the diseased area based on the point cloud data of the diseased area and the location corresponding to the point cloud data of the diseased area;

[0033] Based on the location of the damaged area and the three-dimensional road surface modeling data of the target road surface area, the location, type, size information and affected area information of the damaged area are determined.

[0034] According to a lightweight three-dimensional detection method suitable for rural roads provided by the present invention, the type of the diseased area includes at least one of cracks, potholes, bumps, subsidence, dislocation, repairs, broken plates and plate corner fractures, the size information of the diseased area includes at least one of the length, width and depth of the diseased area, and the affected area information of the diseased area includes at least one of the affected length, affected width, affected depth, affected area and affected degree.

[0035] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, after performing road surface detection on the target road area based on the three-dimensional modeling data of the target road area and the determined road type, and obtaining the detection result, the method further includes:

[0036] Sending the detection results of the road target area to the data management platform;

[0037] The functions of the data management platform include at least one of data analysis, data storage, data distribution and data presentation.

[0038] According to a lightweight three-dimensional detection method applicable to rural roads provided by the present invention, the line scanning three-dimensional measurement sensor includes:

[0039] A sensor comprising a laser and a high-speed 3D camera, wherein the laser is used to project a laser beam vertically onto the road surface, and the high-speed 3D camera is positioned at a certain angle to the laser to obtain cross-sectional data corresponding to the location of the laser line;

[0040] A controller, configured to control the sensor to acquire road surface cross-sectional data;

[0041] The cross-section data includes the elevation and grayscale of the road surface corresponding to the laser line;

[0042] The cross-section data constitute original rural highway pavement three-dimensional contour data along the time sequence of collection.

[0043] The present invention also provides a lightweight three-dimensional detection device suitable for rural roads, comprising a carrier, a line scanning three-dimensional measurement sensor, a positioning device, a processor and a memory; the line scanning three-dimensional measurement sensor, the positioning device and the memory are all electrically connected to the processor;

[0044] The positioning device is used to obtain the position information of the line scanning three-dimensional measurement sensor, and the carrier is used to carry the line scanning three-dimensional measurement sensor to collect point cloud data of the target road section according to the target trajectory;

[0045] It also includes a program or instruction stored in the memory and executable on the processor, and when the program or instruction is executed by the processor, a lightweight three-dimensional detection method applicable to rural roads as described in any of the above is executed.

[0046] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements any of the above-described lightweight three-dimensional detection methods applicable to rural roads.

[0047] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the computer program implements any of the above-described lightweight three-dimensional detection methods applicable to rural roads.

[0048] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described lightweight three-dimensional detection methods applicable to rural roads.

[0049] The lightweight three-dimensional detection method and device for rural roads provided by the present invention utilize a line scanning three-dimensional measurement sensor to obtain the three-dimensional contour data of the road, and then determine the actual road surface target area data and position based on the various characteristics of the road, and then establish the three-dimensional modeling data of the road surface target area to realize the detection of road surface diseases, etc. The detection operation is convenient and the results are accurate, which is suitable for large-scale inspection scenarios of rural roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 This is a schematic flow chart of a lightweight three-dimensional detection method applicable to rural roads provided by the present invention;

[0052] Figure 2 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0054] The following combination Figure 1-Figure 2 The present invention describes a lightweight three-dimensional detection method and device suitable for rural roads.

[0055] The lightweight 3D inspection method for rural roads according to an embodiment of the present invention can be executed by a server. In some embodiments, the execution entity can also be a processor. The specific type of execution entity is not limited here. The following describes the lightweight 3D inspection method for rural roads according to an embodiment of the present invention using a processor as the execution entity.

[0056] Reference Figure 1 The lightweight three-dimensional detection method for rural roads according to an embodiment of the present invention mainly includes steps 110, 120, 130, 140 and 150.

[0057] Step 110: Acquire original rural highway pavement 3D profile data using a line scanning 3D measurement sensor.

[0058] It should be noted that a line-scan 3D measurement sensor is a surveying and mapping device that utilizes 3D sensing technology. It consists of a sensor and a controller. The sensor includes a laser and a high-speed 3D camera. The laser projects a laser beam vertically onto the road surface, while the high-speed 3D camera is positioned at a certain angle (4° to 10°) to the laser to acquire cross-sectional data corresponding to the location of the laser line.

[0059] In some embodiments, the controller may be used to control the sensor to obtain cross-sectional data of the road surface.

[0060] The cross-section data includes the elevation and grayscale of the road surface corresponding to the laser line. The cross-section data constitute the original rural highway road surface 3D contour data along the time sequence of acquisition, that is, the original rural highway road surface 3D contour data is time series data arranged according to the acquisition time.

[0061] It is understandable that after receiving the detection signal, the controller can control the laser to project a laser beam onto the road surface, and control the high-speed 3D camera to collect cross-sectional data corresponding to the position of the laser line.

[0062] It should be noted that the laser beam can be a straight laser line or a multi-line laser beam, which is not limited here.

[0063] In this implementation, a line laser is used to project a one-dimensional laser line onto the road surface. A high-speed 3D camera is positioned at a certain angle (4° to 10°) to the laser to capture cross-sectional data corresponding to the laser line's location. Triangulation is then used to obtain three-dimensional information about the corresponding surface location, creating point cloud data of the road surface. Cross-sectional data is obtained by capturing the road surface profile at multiple measurement points along the road width using the 3D camera at the same measurement position. Over 1,000 measurement points can be located along the road width.

[0064] During the data collection process, data collection does not rely on the external lighting environment, has low requirements on the detection environment, is simple and intuitive to operate, and has a high degree of automation.

[0065] When using a line scanning 3D measurement sensor for detection, the line scanning 3D measurement sensor can be installed on a vehicle, and detection data of multiple positions of the rural road surface can be obtained by moving the vehicle, thereby obtaining the original rural road surface 3D contour data, which is point cloud data.

[0066] It should be noted that rural roads have the characteristics of poor traffic capacity, large variations in road width, complex terrain, and large differences in road conditions. During detection, the original rural highway pavement 3D contour data will also include point cloud data of irrelevant areas such as non-pavement areas.

[0067] Step 120, based on the original rural highway pavement three-dimensional profile data, the pavement target area data and position are extracted by utilizing the macro-continuity characteristics of the pavement surface profile, the characteristics of the non-pavement area being located on both sides of the road width direction, the elevation mutation characteristics of the adjacent positions of the pavement area and the non-pavement area, and the continuity characteristics of the measurement posture.

[0068] It should be noted that the road surface target area is the road surface area of ​​a rural highway. In this case, the road surface target area data and location representing the road surface are extracted from the original rural highway road surface 3D contour data based on the elevation difference characteristics of the adjacent positions of the road surface area and the non-road surface area.

[0069] In some embodiments, based on the original rural highway pavement three-dimensional profile data, for all measurement points in any cross section, the elevation difference between two measurement points with a first preset distance between the measurement points is calculated from the inside to the outside.

[0070] In this embodiment, the inner side refers to a position close to the center of the road surface, and the outer side refers to a position toward the edge of the road surface width.

[0071] It can be understood that based on the characteristics of the non-pavement area being located on both sides of the road width and the elevation mutation characteristics of the adjacent positions of the road surface area and the non-pavement area, the measurement points of the elevation mutation are first determined in the width direction of the road surface.

[0072] The first preset value can be set based on experience, and for road sections in different areas, the first preset value can be set to different values.

[0073] If the elevation difference is greater than the first preset threshold, the measurement point located on the outside will be marked as an abnormal mutation point, which will facilitate further determination of the boundary of the road surface to determine the road surface area, and then determine the position of each point in the road surface area based on the continuity characteristics of the measurement posture and the macro-continuity characteristics of the road surface apparent contour.

[0074] On this basis, a binary map can be generated based on each abnormal mutation point. The binary map includes each abnormal map mutation point. A binary map, also known as a binary image, refers to an image in which each pixel is either black or white.

[0075] In this embodiment, the binary image may be extended using a morphological processing method, and denoising may be performed based on the area and length characteristics of the connected regions in the extended binary image to determine the target area data and location on the road surface.

[0076] It is understandable that the normal area boundary can be found by searching from the inside to the outside according to the macro-continuity characteristics of the road surface profile, so as to confirm the normal road area in the current section and obtain the road target area data and position.

[0077] In this embodiment, by determining the abnormal mutation point, a binary image can be generated to determine the boundary of each cross section out of the road surface, thereby determining the data and position of the target area of ​​the road surface.

[0078] Step 130 : extracting measurement posture information based on the road surface target area data, and eliminating the influence of the measurement posture from the original rural highway road surface 3D contour data to obtain 3D modeling data of the road surface target area.

[0079] It is understandable that during the detection process, due to the unevenness of the road surface, the slope of the road surface and the vibration of the vehicle during driving, the measurement posture of the line scanning three-dimensional measurement sensor fixed to the vehicle will change, and the original rural road surface three-dimensional contour data obtained by detection includes the influence of the measurement posture.

[0080] The road surface model obtained by modeling after determining the target road surface area data does not conform to the actual road surface, so the influence of the measurement posture needs to be eliminated.

[0081] In some embodiments, the difference between the average elevation of the starting cross-section and the average elevation of the ending cross-section corresponding to any cross-section in the road target area data can be first obtained, and the cross-section with a difference greater than a preset threshold can be determined as a jump surface, so as to process the road target area according to all jump surfaces and obtain all divided road sections.

[0082] It should be noted that the starting cross-section is the corresponding cross-section taken forward with the cross-section as the center plane, along the data collection sequence, and at intervals of the first preset length; the ending cross-section is the corresponding cross-section taken backward with the cross-section as the center plane, along the data collection sequence, and at intervals of the first preset length.

[0083] It can be understood that by determining each jump surface according to the average elevation of the cross section, it is convenient to determine the position of the measurement posture change.

[0084] On this basis, the first time series data can be constructed according to the average elevation of all cross sections of each divided road section, and the first time series data can be converted into first frequency data to obtain the pitch and vibration elevation of the corresponding measurement posture of each cross section according to the first frequency data.

[0085] In this embodiment, after processing the first time series data using Fourier transform, first frequency data may be acquired to determine the amplitude and phase information of each spectral component according to the first frequency data.

[0086] Specifically, after obtaining the vibration and pitch frequency range of the measurement posture based on the amplitude and phase information of each spectral component, the pitch and vibration periodic signals of the measurement sensor can be reconstructed according to the inverse Fourier transform and the vibration and pitch frequency range of the measurement posture to obtain the pitch and vibration elevation of the measurement posture corresponding to each cross-section.

[0087] On this basis, second time series data is constructed according to the slopes of all cross sections of each divided road section, and the second time series data is converted into second frequency data to obtain the roll angle of the corresponding measurement posture of each cross section according to the second frequency data.

[0088] In this embodiment, all cross-sections of each divided road section can be processed according to a linear fitting algorithm to obtain the cross-sectional slope corresponding to each cross-section, and the roll angle of each cross-section can be determined based on the cross-sectional slope of each cross-section to construct a second time series data based on the roll angles of all cross-sections.

[0089] Specifically, after processing the second time series data using Fourier transform, the second frequency data can be obtained to determine the amplitude and phase information of each spectral component based on the second frequency data. Then, after obtaining the roll frequency range of the measuring attitude of the measuring sensor based on the amplitude and phase information of each spectral component, the roll periodic signal of the measuring sensor is reconstructed based on the inverse Fourier transform and the roll frequency range of the measuring attitude of the measuring sensor to obtain the roll angle of the measuring attitude corresponding to each cross-section.

[0090] After solving the data of all cross sections, the road target area data can be corrected according to the pitch and vibration elevation of the corresponding measurement posture of the cross section and the roll angle of the corresponding measurement posture of the cross section to obtain the three-dimensional modeling data of the road target area.

[0091] The 3D modeling data of the target road area does not include the influence of the measurement posture. The model built based on this data can accurately reflect the actual 3D contour of the road surface.

[0092] Step 140 : Determine the type of the road surface based on the three-dimensional modeling data of the target road surface area.

[0093] After obtaining the three-dimensional modeling data of the target road area, the data can be further analyzed to determine the type of road surface.

[0094] It is understandable that the types of road surfaces include at least one of asphalt road surface, cement road surface and gravel road surface. Different types of road surfaces have different surface textures.

[0095] In some embodiments, the three-dimensional modeling data of the road target area can be used to calculate the construction depth of each measurement point in the road target area.

[0096] The texture depth of a pavement surface, also known as grain depth, is an important indicator of pavement roughness. It refers to the average depth of open pores within a given area of ​​uneven surface. It is primarily used to assess the pavement's macro-roughness, drainage performance, and skid resistance.

[0097] However, national standards specify construction depth requirements for different types of pavement. For example, according to the "Highway Asphalt Pavement Design Specification JTG D50-2006" and the "Technical Guidelines for Surface Penetration Regeneration and Repair of Asphalt Pavements," the construction depth of asphalt pavements must be no less than 0.55mm. The construction depth of cement concrete pavements is specified as follows: for expressways and first-class highways, the construction depth must be no less than 0.7mm and no more than 1.1mm; for other highways, the construction depth must be no less than 0.5mm and no more than 1.0mm.

[0098] On this basis, the type of road surface can be determined according to the structural depth of the road surface.

[0099] In other words, the road surface texture characteristics of the road surface target area can be obtained according to the structural depth of each measurement point in the road surface target area.

[0100] Considering that some road surface areas may have a certain degree of randomness in data due to various factors, the road surface texture characteristics can be determined based on the structural depth. The road surface texture characteristics include the uniformity, periodicity, and size of the road surface texture distribution. In other words, the uniformity, periodicity, and size of the open pores on the road surface are determined.

[0101] On this basis, by comprehensively considering the characteristics of the open gaps on the entire road surface, the type of road surface can be determined based on the road surface texture characteristics of the target area.

[0102] In this embodiment, the type of the road surface can be conveniently determined by measuring the structural depth of the road surface, providing a simpler method for determining the road surface type.

[0103] Step 150 : Based on the three-dimensional modeling data of the target road surface area and the determined road surface type, a road surface detection is performed on the target road surface area to obtain a detection result.

[0104] The road surface detection includes at least one of road surface damage detection, road surface flatness detection and road surface rutting detection.

[0105] Each different type of road surface has its own common types of defects and patterns. Defects can be identified for each type of road surface, thereby improving the efficiency and accuracy of defect detection.

[0106] In this embodiment, a line scanning three-dimensional measurement sensor is used to obtain the three-dimensional contour data of the highway, and then the actual road surface target area data and position are determined based on the various characteristics of the highway. Then, three-dimensional modeling data of the road surface target area is established to realize the detection of road surface diseases, etc. The detection operation is convenient and the results are accurate, which is suitable for large-scale detection scenarios of rural roads.

[0107] In some embodiments, road surface damage detection may include the following steps.

[0108] The 3D modeling data of the target road area is input into the defect recognition model, which then outputs point cloud data of the defect area. The defect recognition model is trained using point cloud data of the defective road surface as samples and point cloud data of the defective area within the defective road surface as labels.

[0109] It is understandable that the disease recognition model can be established based on the point cloud model PointNet++ or CNN (Convolutional Neural Network).

[0110] Taking the point cloud model as an example, after the 3D modeling data of the target road area is input into the defect recognition model, the PointNet++ model will first sample and divide the 3D modeling data of the target road area into regions, and then perform feature extraction through the PointNet network in each local area. After continuous iteration, the PointNet network is used to extract the global features of the 3D modeling data of the target road area, and then the points in the point cloud are locally divided, the overall features of the local area are extracted, and it is determined whether there are defects in the local area. Finally, the point cloud data of the diseased area is output.

[0111] It is understandable that different disease recognition models can be trained for different types of road surfaces to improve the efficiency and accuracy of identifying different types of road surface diseases.

[0112] On this basis, the location of the diseased area can be determined based on the point cloud data of the diseased area and the corresponding location of the point cloud data of the diseased area. After that, the location of the diseased area can be marked on the map to facilitate subsequent disease review and maintenance.

[0113] In other words, the location, type, size information and affected area information of the defect area can be determined based on the location of the defect area and the road surface target area three-dimensional modeling data of the road surface target area.

[0114] The types of the damaged areas include at least one of cracks, potholes, bumps, subsidence, dislocations, repairs, broken plates and plate corner fractures; the size information of the damaged areas includes at least one of the length, width and depth of the damaged areas; the affected area information of the damaged areas includes at least one of the affected length, affected width, affected depth, affected area and affected degree.

[0115] In this embodiment, by identifying the damaged area, the damage detection of the road surface is achieved, and it is convenient to carry out highway maintenance based on the detection results.

[0116] In some embodiments, road surface roughness detection may include the following steps.

[0117] First, the longitudinal profile data of the preset position can be determined based on the three-dimensional modeling data of the target road area.

[0118] It is understandable that the preset position is the position for performing flatness detection. In this embodiment, it can be selected according to actual conditions and detection requirements.

[0119] The longitudinal profile data may be determined based on the three-dimensional modeling data of the target road area and the elevation data of the road surface along the length direction of the road surface at a preset position.

[0120] In this embodiment, elevation data may be determined in multiple longitudinal directions along the road surface to obtain the flatness.

[0121] In some embodiments, when the preset position is located within a target area of ​​the road surface, an International Roughness Index (IRI) is calculated based on the longitudinal profile data of the preset position.

[0122] In other embodiments, since the preset position is selected in advance and the road surface target area has not yet been determined, if the preset position is not within the road surface target area, then since the preset position is not within the road surface area, only the preset position needs to be marked without calculating the International Roughness Index (IRI).

[0123] In some embodiments, based on the 3D modeling data of the target road surface area and set parameters, the cross-sectional profile (i.e., the road width profile) at the corresponding location can be selected to calculate rutting depth using the envelope method or the three-meter ruler method. If the width of the normal road surface area is less than 3 meters, the envelope method can be used to calculate rutting depth.

[0124] After performing road surface detection on the target road area based on the three-dimensional modeling data of the target road area and the determined road surface type and obtaining the detection results, the lightweight three-dimensional detection method applicable to rural roads in an embodiment of the present invention also includes: sending the detection results of the target road area to a data management platform.

[0125] The functions of the data management platform include at least one of data analysis, data storage, data distribution and data presentation.

[0126] In this case, the real-time collected data can be sent to the data management platform, and the analyzed test results can also be sent to the data management platform. The data management platform can store, analyze and process the data and send it to other terminal devices or servers for subsequent processing.

[0127] The lightweight three-dimensional detection device for rural roads provided by the present invention is described below. The lightweight three-dimensional detection device for rural roads described below and the lightweight three-dimensional detection method for rural roads described above can be referenced to each other.

[0128] The lightweight three-dimensional detection device suitable for rural roads of an embodiment of the present invention includes a carrier, a line scanning three-dimensional measurement sensor, a positioning device, a processor and a memory; the line scanning three-dimensional measurement sensor, the positioning device and the memory are all electrically connected to the processor.

[0129] Vehicles can be any type of car, including sedans, SVUs, pickup trucks, and minivans.

[0130] The positioning device is used to obtain the position information of the line scanning three-dimensional measurement sensor, and the vehicle is used to carry the line scanning three-dimensional measurement sensor to collect point cloud data of the target road section according to the target trajectory.

[0131] The lightweight three-dimensional detection device suitable for rural roads in an embodiment of the present invention also includes a program or instruction stored in a memory and executable on a processor. When the program or instruction is executed by the processor, the lightweight three-dimensional detection method suitable for rural roads as described above is executed.

[0132] According to an embodiment of the present invention, a lightweight three-dimensional inspection device for rural roads is provided. By using a line scanning three-dimensional measurement sensor to obtain three-dimensional contour data of the road, the actual road surface target area data and position are determined based on various characteristics of the road. Then, three-dimensional modeling data of the road surface target area is established to realize the detection of road surface diseases, etc. The inspection operation is convenient and the results are accurate. It is suitable for large-scale inspection scenarios of rural roads.

[0133] Figure 2 An example of a physical structure diagram of an electronic device is shown below. Figure 2 As shown, the electronic device may include: a processor (processor) 210, a communication interface (Communications Interface) 220, a memory (memory) 230 and a communication bus 240, wherein the processor 210, the communication interface 220, and the memory 230 communicate with each other through the communication bus 240. The processor 210 can call the logic instructions in the memory 230 to execute a lightweight three-dimensional detection method suitable for rural roads, the method including: obtaining original rural road pavement three-dimensional contour data using a line scanning three-dimensional measurement sensor; the original rural road pavement three-dimensional contour data is point cloud data; based on the original rural road pavement three-dimensional contour data, the pavement target area data and position are extracted using the macro continuity characteristics of the pavement surface contour, the characteristics of the non-pavement area being located on both sides of the road width direction, the elevation mutation characteristics of the adjacent positions of the pavement area and the non-pavement area, and the continuity characteristics of the measurement posture; based on the pavement target area data, the measurement posture information is extracted, and the measurement posture influence is eliminated from the original rural road pavement three-dimensional contour data to obtain pavement target area three-dimensional modeling data; based on the pavement target area three-dimensional modeling data, the type of pavement is determined; the type of pavement includes at least one of asphalt pavement, cement pavement, and gravel pavement; based on the pavement target area three-dimensional modeling data and the determined pavement type, the pavement target area is subjected to pavement detection to obtain a detection result; the pavement detection includes at least one of pavement damage detection, pavement flatness detection, and pavement rutting detection.

[0134] In addition, the logic instructions in the above-mentioned memory 230 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0135] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the lightweight three-dimensional detection method for rural roads provided by the above methods, the method comprising: using a line scanning three-dimensional measurement sensor to obtain original rural road pavement three-dimensional contour data; the original rural road pavement three-dimensional contour data is point cloud data; based on the original rural road pavement three-dimensional contour data, using the macro-continuity characteristics of the pavement surface contour, the characteristics of the non-pavement area on both sides of the road width direction, the characteristics of the pavement area and the non-pavement area relative to each other, The elevation mutation characteristics of adjacent positions and the continuity characteristics of the measurement posture are used to extract the pavement target area data and position; based on the pavement target area data, the measurement posture information is extracted, and the influence of the measurement posture is eliminated from the original rural road pavement 3D contour data to obtain the pavement target area 3D modeling data; based on the pavement target area 3D modeling data, the type of pavement is determined; the type of pavement includes at least one of asphalt pavement, cement pavement, and gravel pavement; based on the pavement target area 3D modeling data and the determined pavement type, the pavement target area is subjected to pavement detection to obtain a detection result; the pavement detection includes at least one of pavement damage detection, pavement flatness detection, and pavement rutting detection.

[0136] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the lightweight three-dimensional detection method for rural roads provided by the above-mentioned methods, the method comprising: obtaining original rural road pavement three-dimensional contour data using a line scanning three-dimensional measurement sensor; the original rural road pavement three-dimensional contour data is point cloud data; based on the original rural road pavement three-dimensional contour data, the pavement target area data and position are extracted using the macro-continuity characteristics of the pavement surface contour, the characteristics of the non-pavement area being located on both sides of the road width direction, the elevation mutation characteristics of the adjacent positions of the pavement area and the non-pavement area, and the continuity characteristics of the measurement posture; based on the pavement target area data, the measurement posture information is extracted, and the measurement posture influence is eliminated from the original rural road pavement three-dimensional contour data to obtain pavement target area three-dimensional modeling data; based on the pavement target area three-dimensional modeling data, the type of pavement is determined; the type of pavement includes at least one of asphalt pavement, cement pavement, and gravel pavement; based on the pavement target area three-dimensional modeling data and the determined pavement type, the pavement target area is subjected to pavement detection to obtain a detection result; the pavement detection includes at least one of pavement damage detection, pavement flatness detection, and pavement rutting detection.

[0137] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0138] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A lightweight three-dimensional detection method suitable for rural roads, characterized in that: include: Use line scanning 3D measurement sensors to obtain original rural road pavement 3D profile data; The original rural road pavement 3D contour data is point cloud data; Based on the original rural highway pavement 3D profile data, the target pavement area data and location are extracted by utilizing the macroscopic continuity characteristics of the pavement surface profile, the characteristics of the non-pavement area being located on both sides of the road width, the elevation mutation characteristics of the adjacent positions of the pavement area and the non-pavement area, and the continuity characteristics of the measurement posture; Based on the road surface target area data, measurement posture information is extracted, and the influence of the measurement posture is eliminated from the original rural highway road surface three-dimensional contour data to obtain three-dimensional modeling data of the road surface target area, including: Obtaining the difference between the average elevation of the starting cross section and the average elevation of the ending cross section corresponding to any cross section in the road target area data, determining the cross section with the difference greater than a preset threshold as a jump surface, processing the road target area according to all jump surfaces, and obtaining all divided road sections; Constructing first time series data based on the average elevation of all cross sections of each divided road section, converting the first time series data into first frequency data, and obtaining the pitch and vibration elevation of the corresponding measurement posture of each cross section based on the first frequency data; Constructing second time series data based on the slopes of all cross sections of each divided road section, converting the second time series data into second frequency data, and obtaining a roll angle corresponding to a measured posture of each cross section based on the second frequency data; Correcting the road target area data according to the pitch and vibration elevation of the cross-section corresponding to the measured posture and the roll angle of the cross-section corresponding to the measured posture to obtain three-dimensional modeling data of the road target area; The starting cross section is a corresponding cross section taken forward at intervals of a first preset length along the data acquisition sequence with the cross section as the center plane; the ending cross section is a corresponding cross section taken backward at intervals of a first preset length along the data acquisition sequence with the cross section as the center plane; Determining a road surface type based on the three-dimensional modeling data of the road surface target area; the road surface type includes at least one of an asphalt road surface, a cement road surface, and a gravel road surface; Based on the three-dimensional modeling data of the target road area and the determined road type, a road surface detection is performed on the target road area to obtain a detection result; the road surface detection includes at least one of road damage detection, road surface flatness detection and road surface rutting detection.

2. The lightweight three-dimensional detection method applicable to rural roads according to claim 1 is characterized in that: The method extracts the target road area data and position based on the original rural road pavement 3D profile data, utilizing the macroscopic continuity feature of the pavement surface profile, the feature that the non-pavement area is located on both sides of the road width, the elevation mutation feature of the adjacent positions of the pavement area and the non-pavement area, and the continuity feature of the measurement posture, including: Based on the original rural highway pavement three-dimensional profile data, for all measurement points in any cross section, the elevation difference between two measurement points with a first preset distance between the measurement points is calculated from the inside to the outside; If the elevation difference is greater than a first preset threshold, the measuring point located outside is marked as an abnormal mutation point; Generate a binary graph based on each abnormal mutation point; The binary image is extended using a morphological processing method, and denoising is performed based on the area and length characteristics of the connected regions in the extended binary image to determine the data and position of the target road area.

3. The lightweight three-dimensional detection method for rural roads according to claim 1 is characterized in that: The determining of the type of the road surface based on the three-dimensional modeling data of the road surface target area includes: Calculating the structural depth of each measuring point in the road surface target area based on the road surface target area three-dimensional modeling data of the road surface target area; Obtaining a road surface texture feature of the target road surface area according to the structural depth of each measurement point in the target road surface area; the road surface texture feature includes uniformity, periodicity, and size data of the road surface texture distribution; The type of the road surface is determined based on the road surface texture characteristics of the target road surface area.

4. The lightweight three-dimensional detection method applicable to rural roads according to claim 1, characterized in that: The road surface smoothness detection comprises the following steps: determining longitudinal profile data of a preset position based on the three-dimensional modeling data of the road surface target area; When the preset position is located within the road surface target area, calculating an International Roughness Index (IRI) based on the longitudinal profile data of the preset position; In a case where the preset position is not located within the road surface target area, the preset position is marked.

5. The lightweight three-dimensional detection method applicable to rural roads according to claim 1, characterized in that: The road surface damage detection comprises the following steps: Inputting the three-dimensional modeling data of the target road surface area into a defect recognition model to obtain point cloud data of the defect area output by the defect recognition model; the defect recognition model is trained using the point cloud data of the defective road surface as a sample and the point cloud data of the defect area in the defective road surface as a label; Determining the location of the diseased area based on the point cloud data of the diseased area and the location corresponding to the point cloud data of the diseased area; Based on the position of the damaged area and the three-dimensional modeling data of the road surface target area, the position, type, size information and affected area information of the damaged area are determined.

6. The lightweight three-dimensional detection method applicable to rural roads according to claim 5, characterized in that: The type of the damaged area includes at least one of cracks, potholes, bulges, subsidence, dislocation, repairs, broken plates and plate corner fractures; the size information of the damaged area includes at least one of the length, width and depth of the damaged area; the affected area information of the damaged area includes at least one of the affected length, affected width, affected depth, affected area and affected degree.

7. The lightweight three-dimensional detection method applicable to rural roads according to claim 1, characterized in that: After performing road surface detection on the road surface target area based on the three-dimensional modeling data of the road surface target area and the determined road surface type to obtain a detection result, the method further includes: Sending the detection results of the road target area to the data management platform; The functions of the data management platform include at least one of data analysis, data storage, data distribution and data presentation.

8. The lightweight three-dimensional detection method applicable to rural roads according to claim 1, characterized in that: The line scanning three-dimensional measurement sensor comprises: A sensor comprising a laser and a high-speed 3D camera, wherein the laser is used to project a laser beam vertically onto the road surface, and the high-speed 3D camera is positioned at a certain angle to the laser to obtain cross-sectional data corresponding to the location of the laser line; A controller, configured to control the sensor to acquire road surface cross-sectional data; The cross-section data includes the elevation and grayscale of the road surface corresponding to the laser line; The cross-section data constitute original rural highway pavement three-dimensional contour data along the time sequence of collection.

9. A lightweight three-dimensional detection device suitable for rural roads, characterized in that: It includes a carrier, a line scanning three-dimensional measurement sensor, a positioning device, a processor and a memory; the line scanning three-dimensional measurement sensor, the positioning device and the memory are all electrically connected to the processor; The positioning device is used to obtain the position information of the line scanning three-dimensional measurement sensor, and the carrier is used to carry the line scanning three-dimensional measurement sensor to collect point cloud data of the target road section according to the target trajectory; it also includes a program or instruction stored in the memory and executable on the processor, and when the program or instruction is executed by the processor, it executes the lightweight three-dimensional detection method applicable to rural roads as described in any one of claims 1 to 8.

Images