CN110084116B - Road surface detection method, road surface detection device, computer equipment and storage medium - Google Patents
Road surface detection method, road surface detection device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110084116B CN110084116B CN201910220802.6A CN201910220802A CN110084116B CN 110084116 B CN110084116 B CN 110084116B CN 201910220802 A CN201910220802 A CN 201910220802A CN 110084116 B CN110084116 B CN 110084116B
- Authority
- CN
- China
- Prior art keywords
- grid
- value
- point cloud
- ground
- ground point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/182—Network patterns, e.g. roads or rivers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Road Repair (AREA)
- Traffic Control Systems (AREA)
Abstract
The application relates to a road surface detection method, a road surface detection device, a computer device and a storage medium. The method comprises the following steps: acquiring ground point cloud of a road surface to be detected; constructing a grid map according to the ground point cloud, and determining two-dimensional ground points covered by each grid in the grid map; and acquiring the height value of the two-dimensional ground points covered by each grid, and determining the flatness of the road surface to be detected according to the height value of the two-dimensional ground points covered by each grid. By adopting the method, the pavement detection with higher precision can be realized.
Description
Technical Field
The present application relates to the field of unmanned driving technologies, and in particular, to a road surface detection method, apparatus, computer device, and storage medium.
Background
Along with the development in the field of unmanned driving, a road surface road condition detection technology appears, a road surface road condition detection module is one of the most basic and important modules for automatic driving, and the road surface road condition detection technology can provide road condition guidance for unmanned vehicles.
However, the current road condition detection method is based on camera data, which is greatly affected by the light intensity, and therefore a road condition detection result with high accuracy cannot be obtained.
Disclosure of Invention
In view of the above, it is desirable to provide a road surface detection method, a road surface detection device, a computer apparatus, and a storage medium capable of obtaining a road surface detection result with higher accuracy.
A method of road surface inspection, the method comprising:
acquiring ground point cloud of a road surface to be detected;
constructing a grid map according to the ground point cloud, and determining two-dimensional ground points covered by each grid in the grid map;
and acquiring the height value of the two-dimensional ground points covered by each grid, and determining the flatness of the road surface to be detected according to the height value of the two-dimensional ground points covered by each grid.
In one embodiment, the acquiring a ground point cloud of a road surface to be detected includes:
acquiring point cloud data of a road surface to be detected, which is acquired by an acquisition device; the collecting device is a laser radar;
and extracting the ground point cloud of the road surface to be detected from the point cloud data.
In one embodiment, the extracting the ground point cloud of the road surface to be detected from the point cloud data includes:
acquiring a preset number of point clouds from the point cloud data according to the distance from the acquisition device to the far;
determining ground point clouds in the preset number of point clouds according to the height of the acquisition device;
and performing straight line fitting according to the determined ground point cloud to determine the ground point cloud in the point cloud data.
In one embodiment, the constructing a grid map according to the ground point cloud, and determining two-dimensional ground points covered by each grid in the grid map comprises:
converting the ground point cloud to be detected into two-dimensional ground points according to the three-dimensional information of the ground point cloud to be detected;
and covering the two-dimensional ground points by adopting a grid map with the grid size of a first preset value to obtain a grid map.
In one embodiment, the obtaining the height value of the two-dimensional ground point covered by each grid, and determining the flatness of the ground to be detected according to the height value of the two-dimensional ground point covered by each grid includes:
acquiring a height value of the two-dimensional ground point covered by each grid;
counting the height value of the two-dimensional ground point covered by each grid to obtain a statistical value of each grid;
constructing a template with the side length of a second preset value by taking each grid as a center, and determining the flatness of each grid according to the statistic value of each grid in the template;
and acquiring the average value or the median value of the flatness of each grid, and taking the average value or the median value of the flatness of each grid as the flatness of the pavement to be detected.
In one embodiment, the obtaining of the height value of the two-dimensional ground point covered by each grid; counting the height value of the two-dimensional ground point covered by each grid to obtain the statistical value of each grid, wherein the statistical value comprises the following steps:
traversing the two-dimensional ground points covered by each grid to obtain the maximum height value or the average height value or the minimum height value of the ground points covered by each grid;
and taking the maximum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid as the corresponding statistical value of each grid.
In one embodiment, after the performing statistical analysis on the height values of the two-dimensional ground points covered by each grid to obtain the statistical values of each grid, the method further includes: and carrying out interpolation processing on the grids with empty statistical values.
In one embodiment, the interpolating the grid whose statistics are empty includes:
constructing a template with the side length of a third preset value by taking the grid with the empty statistical value as a center;
and acquiring the mean value of the statistical value of each grid in the constructed template, and taking the mean value of the statistical value as the interpolation of the grid with the empty statistical value.
In one embodiment, the road surface detection method further includes:
and carrying out pit detection or fault detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid.
A road surface detecting device, the device comprising:
the ground point cloud acquisition module is used for acquiring ground point clouds to be detected;
the grid map building module is used for building a grid map according to the ground point clouds and determining two-dimensional ground points covered by each grid in the grid map;
and the detection analysis module is used for acquiring the height value of the two-dimensional ground point covered by each grid, determining the flatness of the road surface to be detected according to the height value of the two-dimensional ground point covered by each grid, and taking the flatness of the road surface to be detected as a detection result.
In one embodiment, the ground point cloud obtaining module comprises:
the acquisition unit is used for acquiring point cloud data of the to-be-detected road surface acquired by the acquisition device; the collecting device is a laser radar;
and the extraction unit is used for extracting the ground point cloud of the road surface to be detected from the point cloud data.
The extraction unit is also used for acquiring a preset number of point clouds from the point cloud data from near to far according to a distance acquisition device, determining ground point clouds in the preset number of point clouds according to the height of the acquisition device, and performing straight line fitting according to the determined ground point clouds to determine the ground point clouds in the point cloud data.
In one embodiment, the grid map building module comprises:
the projection unit is used for converting the ground point cloud to be detected into two-dimensional ground points according to the three-dimensional information of the ground point cloud to be detected;
and the grid dividing unit is used for covering the two-dimensional ground points by adopting a grid map with the grid size of a first preset value to obtain a grid map.
In one embodiment, the detection analysis module comprises:
the data acquisition unit is used for acquiring the height value of the two-dimensional ground point covered by each grid;
the statistical unit is used for carrying out statistics on the height value of the two-dimensional ground point covered by each grid to obtain the statistical value of each grid;
the first calculation unit is used for constructing a template with the side length as a second preset value by taking each grid as a center, and determining the flatness of each grid according to the statistic value of each grid in the template;
and the second calculating unit is used for acquiring the average value or the median value of the flatness of each grid, and taking the average value or the median value of the flatness of each grid as the flatness of the road surface to be detected.
In one embodiment, the statistical unit is further configured to traverse the two-dimensional ground points covered by each grid to obtain a maximum height value, an average height value, or a minimum height value of the two-dimensional ground points covered by each grid, and use the maximum height value, the average height value, or the minimum height value of the two-dimensional ground points covered by each grid as the corresponding statistical value of each grid.
In one embodiment, the detection analysis module further comprises: and the interpolation processing unit is used for carrying out interpolation processing on the grids with the empty statistical values.
In one embodiment, the interpolation processing unit is further configured to construct a template with a side length of a third preset value by taking the grid with the empty statistical value as a center; and acquiring the mean value of the statistical value of each grid in the constructed template, and taking the mean value of the statistical value as the interpolation of the grid with the empty statistical value.
In one embodiment, the road surface detection device may further include:
the pit detection module is used for carrying out pit detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid;
and the fault detection module is used for carrying out fault detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid.
A computer device comprising a memory storing a computer program and a processor implementing the road surface detection step described above when the processor executes the computer program.
A computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the above-described road surface detecting step.
According to the road surface detection method, the device, the computer equipment and the storage medium, the three-dimensional model data reflecting the road surface to be detected is obtained by collecting the ground point cloud data, the grid map is built according to the ground point cloud, the two-dimensional ground points covered by each grid in the grid map are determined, the flatness of the road surface to be detected is obtained by analyzing and processing the height values of the two-dimensional ground points covered by each grid, and the road surface detection with higher precision is realized because the point cloud data is not influenced by light rays and the collected data is accurate, so that the detected flatness precision is high.
Drawings
FIG. 1 is a diagram of an exemplary environment in which a method for detecting a roadway surface is implemented;
FIG. 2 is a schematic flow chart of a method for detecting a road surface according to one embodiment;
FIG. 3 is a schematic diagram of line fitting in one embodiment;
FIG. 4 is a schematic flow chart of the ground point cloud acquisition step in one embodiment;
FIG. 5 is a flowchart illustrating the grid map construction step in one embodiment;
FIG. 6 is a schematic flow chart of the detection analysis step in one embodiment;
FIG. 7a is a grid map without interpolation in one embodiment;
FIG. 7b is a grid map after interpolation processing in one embodiment;
FIG. 8a is a grid map constructed in another embodiment;
FIG. 8b is a grid map after interpolation processing in another embodiment;
FIG. 9 is a schematic flow chart of a road surface detection method in another embodiment;
FIG. 10 is a top view of a two-dimensional ground point in one embodiment;
fig. 11 is a block diagram showing the structure of a road surface detecting device in one embodiment;
FIG. 12 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Fig. 1 is an application environment diagram of a road surface detection method in one embodiment. The embodiment of the application provides a road surface detection method which can be applied to the application environment shown in fig. 1. Wherein, the mobile trolley is provided with a computer device 100 and a collecting device 102. The collection device 102 may be a laser radar, and may emit laser for scanning to obtain point cloud data of the surrounding environment. As shown in fig. 1, the collecting device 102 emits laser onto the road surface to scan the point cloud data of all scanned objects, such as the ground and obstacles, wherein the ground may include road surface features, such as water pits. The acquisition means 102 then transmit the point cloud data of all scanned objects to the computer device 100.
The acquisition device 102 may also be a depth camera for acquiring a depth image of the surrounding environment, and the depth image may be processed and converted into a point cloud image.
The computer device 100 may be a vehicle-mounted terminal, a desktop terminal, or a mobile terminal, and the mobile terminal may be a mobile phone, a tablet computer, a notebook computer, a wearable device, a personal digital assistant, or the like. The computer device 100 may also be implemented as a stand-alone server or as a server cluster comprising a plurality of servers.
Fig. 2 is a schematic flow chart of a road surface detection method in one embodiment. As shown in fig. 2, a road surface detection method is provided, which is described by taking the method as an example applied to the computer device in fig. 1, and comprises the following steps:
The ground point cloud may refer to point cloud data of a ground portion extracted from the point cloud data of the surrounding environment of the vehicle.
Specifically, the acquisition device 102 acquires point cloud data of the vehicle surrounding environment and transmits the point cloud data to the computer device 100, and the computer device 100 performs ground point detection on the acquired point cloud data of the vehicle surrounding environment to extract ground point cloud.
Optionally, the process of acquiring the ground point cloud may also be: the acquisition device 102 acquires a depth image of the surrounding environment of the vehicle and transmits the depth image to the computer device 100, the computer device 100 converts the acquired depth image to obtain point cloud data of the surrounding environment, then ground point detection is performed on the acquired point cloud data of the surrounding environment of the vehicle, and ground point cloud is extracted.
Optionally, the above ground point detection method may be a line-fit algorithm, a point net algorithm, or a detection algorithm based on deep learning.
Referring to an application environment diagram of the road surface detection method in the embodiment of fig. 1 and a schematic diagram of line fitting in the embodiment of fig. 3, a method for performing ground point detection according to a line-fit algorithm includes: judging two point clouds close to the moving trolley as ground point clouds according to the contact height of the laser radar off-vehicle tire and the ground, and identifying the ground point clouds; and performing linear fitting and phase extension processing according to the identified ground point cloud and the detected other point clouds, and judging whether the detected other point clouds are the ground point clouds. And repeating the operation to detect all ground point clouds.
The detection algorithm based on deep learning can be to collect ground point cloud in advance, and perform characteristic comparison according to the collected ground point cloud and point cloud data of a road surface to be detected to obtain the ground point cloud.
And 204, constructing a grid map according to the ground point clouds, and determining two-dimensional ground points covered by each grid in the grid map.
The two-dimensional ground point can be a two-dimensional point formed by projecting ground point cloud in a three-dimensional state onto a plane, and the two-dimensional ground point comprises two-dimensional coordinate information and height information. The grid map may refer to a map including height information and position information of the ground to be detected.
Specifically, the computer device 100 projects a three-dimensional ground point cloud onto a plane to obtain two-dimensional ground points, and then divides the plane including the two-dimensional ground points into grids of a first preset value to obtain a grid map. When the ground point cloud in the three-dimensional state is converted into the two-dimensional ground point, the height value of the ground point cloud in the three-dimensional state in the Z-axis direction is reserved and used as the height value of the two-dimensional ground point.
And step 206, acquiring the height value of the two-dimensional ground points covered by each grid, and determining the flatness of the road surface to be detected according to the height value of the two-dimensional ground points covered by each grid.
The number of the two-dimensional ground points covered by each grid may be zero, or may be one or more than two.
Specifically, the computer device 100 obtains a statistical value of each grid according to the height value of the two-dimensional ground point in each grid, and performs the flatness detection on the road surface to be detected according to the statistical value of each grid. Wherein the flatness detection may include: and calculating the variance of the statistical values of the grids in the preset range around the central grid according to the statistical values of the grids in the preset range around the central grid by taking a certain grid as the central grid, wherein the larger the variance is, the lower the flatness is. Optionally, each grid in the preset area may be set as a central grid, each central grid is traversed, the operation is repeated, a mean value, a median value, a maximum value or a minimum value of statistical values of the grids in a predetermined range around each central grid is calculated, and then a variance is calculated according to the mean value, the median value, the maximum value or the minimum value and is used as the flatness of the preset area.
According to the pavement detection method, the three-dimensional model data reflecting the pavement to be detected is obtained by collecting the ground point cloud data, the grid map is constructed according to the ground point cloud, the two-dimensional ground points covered by each grid in the grid map are determined, the flatness of the pavement to be detected is obtained by analyzing and processing the height values of the two-dimensional ground points covered by each grid, and the pavement detection with higher precision is realized because the point cloud data is not influenced by light rays and the collected data is accurate, so that the detected flatness precision is high.
FIG. 4 is a flowchart illustrating the ground point cloud obtaining step according to an embodiment. As shown in FIG. 4, in one embodiment, step 202 includes the steps of:
And step 404, extracting the ground point cloud of the road surface to be detected from the point cloud data.
Specifically, the computer device 100 performs ground point detection according to the point cloud data to obtain a ground point cloud of the road surface to be detected.
In the above-mentioned ground point cloud acquires the step, through gathering the point cloud data of vehicle surrounding environment, draws ground point cloud from this point cloud data again, has realized waiting to detect the effective acquisition of road surface point cloud, compares the camera data that is changeed the light influence among the data acquisition process, and point cloud data can reflect the actual three-dimensional model information of waiting to detect the road surface more steadily for the road surface testing result accuracy is higher.
In one embodiment, extracting the ground point cloud of the road surface to be detected from the point cloud data includes: acquiring a preset number of point clouds from the point cloud data according to the distance from the acquisition device to the far; determining ground point clouds in the preset number of point clouds according to the height of the acquisition device; and performing straight line fitting according to the determined ground point cloud to determine the ground point cloud in the point cloud data.
The acquisition device acquires a frame of point cloud data, and each point cloud in the frame of point cloud data is generated by a coordinate system established by taking the acquisition device as an origin.
Specifically, referring to fig. 1, the acquisition device 102 first acquires two point clouds closest thereto. And as the first point cloud and the second point cloud, respectively comparing a height value of the first point cloud and a height value of the second point cloud by using a vertical distance between a contact point of the tire and the ground and the acquisition device 102 as a height reference value, wherein if a difference value between the height reference value and the height value of the first point cloud is within a preset interval, the first point cloud is the ground point cloud, and if the difference value between the height reference value and the height value of the second point cloud is within the preset interval, the second point cloud is the ground point cloud. The preset number of point clouds are sequentially acquired from near to far according to the distance acquisition device 102, and the judgment operation is repeated until two point clouds, namely a first ground point cloud and a second ground point cloud, appear. And performing straight line fitting according to the first ground point cloud and the second ground point cloud, and judging whether the subsequently acquired point cloud is the ground point cloud.
In the ground point detection step, the effective extraction of the ground point cloud is realized by a straight line fitting method.
FIG. 5 is a flowchart illustrating the grid map construction step in one embodiment. As shown in FIG. 5, in one embodiment, step 204 includes the steps of:
The ground point cloud comprises three-dimensional coordinate information, and the two-dimensional ground point comprises two-dimensional coordinate information and height information.
And 504, covering the two-dimensional ground points by using a grid map with the grid size of a first preset value to obtain a grid map.
The grid with the size of the first preset value may refer to a grid with a side length of a preset value, for example, a grid with a size of 0.3m by 0.3 m.
In the step of constructing the grid map, a two-dimensional ground point containing a height value is formed by projecting the ground point cloud in a three-dimensional state, and grids are arranged to cover the two-dimensional ground point, so that the grid map is obtained, wherein each grid contains the height value of the two-dimensional ground point covered by the grid. In this embodiment, the grid map is constructed to facilitate detection and analysis according to information contained in each grid.
FIG. 6 is a flow diagram illustrating the steps of the detection analysis in one embodiment. As shown in FIG. 6, in one embodiment, step 206 includes the steps of:
The statistical value of the grid can be the assignment of the grid calculated according to the height value of the two-dimensional ground points covered by the grid.
And 606, constructing a template with the side length of a second preset value by taking each grid as a center, and determining the flatness of each grid according to the statistic value of each grid in the template.
The flatness of the grid can reflect the flatness of the road surface area under the world coordinate system corresponding to the grid.
And the flatness of the road surface to be detected is the detection result of the road surface to be detected.
In the detection and analysis steps, the flatness of the road surface to be detected is determined through the height value of the two-dimensional ground points covered by each grid, so that the effective detection of the flatness of the road surface to be detected is realized, wherein the height value of the two-dimensional ground points covered by each grid is counted, the statistical value is assigned to the grid, and then the statistical value of each grid is analyzed, so that the detection result of the road surface detection is closer to the actual condition.
In an embodiment, the obtaining the height value of the two-dimensional ground point covered by each grid, and performing statistics on the height value of the two-dimensional ground point covered by each grid to obtain a statistical value of each grid includes: traversing the two-dimensional ground points covered by each grid to obtain the maximum height value or the average height value of the two-dimensional ground points covered by each grid; and taking the maximum height value or the average value of the height values of the two-dimensional ground points covered by each grid as the corresponding statistical value of each grid.
For example, in a certain grid, two-dimensional ground points 1 and two-dimensional ground points 2 are covered, wherein the height value of the two-dimensional ground point 1 is 1.1, the height value of the two-dimensional ground point 2 is 1.3, the maximum height value 1.3 can be taken as the statistical value of the grid, and the average value of the height values 1.2 can also be taken as the statistical value of the grid.
Optionally, the minimum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid may also be used as the corresponding statistical value of each grid.
In the step of obtaining the statistical value of each grid, the statistical value of each grid is determined by the height value of the two-dimensional ground points covered by each grid, and the height information contained in the two-dimensional ground points covered by each grid is converted to the grid corresponding to the two-dimensional ground points, so that the detection and analysis processes do not need to traverse all the two-dimensional ground points, and the analysis is performed according to the height information contained in each grid.
In an embodiment, after performing statistical analysis on the height values of the two-dimensional ground points covered by each grid to obtain the statistical values of each grid, the method further includes: and carrying out interpolation processing on the grids with empty statistical values.
The interpolation processing refers to assigning a value to a grid whose statistical value is empty.
For example, referring to fig. 7a and 7b, fig. 7a is a grid map without interpolation in one embodiment, and fig. 7b is a grid map after interpolation processing in one embodiment, and as can be seen from fig. 7a and 7b, the grid map with sparse point clouds becomes a grid map with dense point clouds through interpolation processing.
In an embodiment, the interpolating the grid whose statistics are empty includes: constructing a template with the side length of a third preset value by taking the grid with the empty statistical value as a center; and acquiring the mean value of the statistical value of each grid in the constructed template, and taking the mean value of the statistical value as the interpolation of the grid with the empty statistical value.
For example, referring to fig. 8a, fig. 8a is a grid map without interpolation processing in another embodiment, and fig. 8b is a grid map after interpolation processing in another embodiment, in a grid at a, two grids of the surrounding grids covered by 3 × 3 templates have statistical values, which are 1.5 and 1.7 respectively, so that point a is assigned with a value of 1.6.
Alternatively, a median or a maximum or a minimum of the statistical values of each grid in the constructed template may be obtained, and the median or the maximum or the minimum may be used as an interpolation of the grid whose statistical value is empty.
The step of interpolating the grid with the empty statistical value determines the interpolation of the grid according to the statistical value of the grid adjacent to the grid with the empty statistical value, wherein the statistical value originally has new assignment to the empty grid, the assignment to the grid can be used for detection and analysis, and the available data of the detection and analysis is more, so that the result of the detection and analysis is more accurate.
In another embodiment, the road surface detection method further includes: and carrying out pit detection or fault detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid.
Optionally, the statistical value of each grid is determined according to the height value of the two-dimensional ground point covered by each grid. And traversing each grid, if the statistical value of the grid is empty, determining that the region under the world coordinate system corresponding to the grid belongs to the fault region, and if the difference value between the statistical value of the grid and the height reference value is greater than a preset value, determining that the region under the world coordinate system corresponding to the grid belongs to the pit region. And judging the region under the world coordinate system corresponding to the continuous grids with the plurality of empty statistical values as a fault region. And judging the area under the world coordinate system corresponding to the continuous grid with the plurality of statistical values and the height reference value having the difference value larger than the preset value as a fault area. The pit area refers to a ground area with low-lying characteristics, and the fault area refers to a ground area with crack characteristics.
Optionally, the statistical value of each grid in the area with the preset size is determined according to the height value of the two-dimensional ground point covered by each grid in the area with the preset size. Scanning each grid in the area with the preset size row by row or column by column, and calculating the average value, the median value, the minimum value or the maximum value of the statistical values of the grids in the area with the preset size as the height value of the area with the preset size. And if the difference between the height value of the area with the preset size and the statistical value of the grid adjacent to the boundary line is larger than a preset value, determining that the area under the world coordinate system corresponding to the area with the preset size is a pit area.
The pit detection step and the fault detection step are used for carrying out pit detection and fault detection according to the height value of the two-dimensional ground point covered by each grid, so that multifunctional road surface detection is realized.
Fig. 9 is a schematic flow chart of another road surface detection method, as shown in fig. 9, in another embodiment, the another road surface detection method includes the following steps:
and 902, performing ground point detection on the collected point cloud data through a ground point algorithm to extract ground point cloud.
Specifically, referring to fig. 1 and 3, fig. 3 is a schematic diagram of straight line fitting. The acquisition device 102 in this embodiment is a multi-line laser radar, and as shown in fig. 1, the multi-line laser radar can simultaneously emit multi-line laser from near to far to scan the surrounding environment, and perform three-dimensional modeling according to the laser point cloud data of the surrounding environment. Ground point detection is then performed. Firstly, according to the height of a laser radar from a tire, judging that two point clouds close to a moving trolley are ground point clouds, and marking the ground point clouds as a first color; and performing linear fitting and phase extension processing according to the ground point cloud marked as the first color and the detected other point clouds, and judging whether the detected other point clouds are the ground point clouds. Wherein the ground point cloud is marked with a first color and the obstacle is marked with a second color. For example, the first color may be green and the second color may be red. And repeating the operation, and screening out all ground point clouds according to the marked first color.
And 904, constructing a grid map according to the ground point cloud.
The ground point cloud comprises three-dimensional position information (x, y, z), wherein an x-y plane is a horizontal plane where the laser radar is located, and a z axis is a long axis perpendicular to the x-y plane.
Specifically, the computer device 100 projects a ground point cloud onto an x-y plane, fig. 10, which is a top view of two-dimensional ground points in one embodiment, wherein the ground point cloud in a three-dimensional state is converted into two-dimensional ground points. And dividing the x-y plane into grids with the size of r, wherein r is a first preset value and can be set according to actual conditions. Referring to fig. 8a, fig. 8a is a constructed grid map, wherein the grid size is 0.1m by 0.1m, and two-dimensional ground points are covered in the grid.
The fact that the grid is empty in statistics means that the grid does not cover the two-dimensional ground points, that is, the height values of the two-dimensional ground points are not obtained.
After the grid map is constructed, some grids are covered with the two-dimensional ground points, and some grids are not covered with the two-dimensional ground points, and at this time, the grids which are not covered with the two-dimensional ground points need to be interpolated, so that a continuous grid map is obtained.
In this embodiment, bilinear interpolation is used to interpolate a grid that does not cover two-dimensional ground points, and for a grid that is far from the grid that covers two-dimensional ground points, interpolation is not performed even if the statistical value of the grid is null. For example, referring to fig. 8b, fig. 8b is a grid map after interpolation processing, and for a grid whose statistics are empty, a 3 × 3 template is constructed; with this grid as the center, in this 3 × 3 template, the surrounding grids are covered, and if the statistical values of the surrounding grids are not empty, the average value of the statistical values of these grids is calculated as the statistical value of the center grid. In fig. 8b, of the surrounding grids covered by the template at grid a, 3 x 3, there are two grids with statistical values of 1.5 and 1.7, respectively, thus assigning a value of 1.6 to point a. At B the grid, whose template of 3 x 3 covers the grid has no statistics, so the grid at B is assigned a null.
And 908, determining the flatness of the road surface to be detected according to the height value of the two-dimensional ground point in the grid.
Specifically, as shown in fig. 8b, a 5 × 5 template is constructed with the grid at point C as the center, the variance is calculated according to the statistical values of all the grids within the template range, the average value is calculated according to the variance, and the average value is used as the flatness of the road surface.
The neighborhood range may refer to a range of a preset interval size, and the neighborhood range is determined according to the height values included in the two-dimensional ground points detected in steps 602 and 604.
According to the pavement detection method, the ground point cloud is extracted, the grid map is constructed, and the grid map is analyzed, so that the flatness, fault judgment and pit judgment of the pavement are obtained, and effective pavement detection is realized.
It should be understood that although the steps in the flowcharts of fig. 2, 4, 5, 6, 9 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2, 4, 5, 6, and 9 may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternatingly with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 11, there is provided a road surface detecting device including: a ground point cloud obtaining module 1102, a grid map building module 1104, and a detection analysis module 1106, wherein:
a ground point cloud obtaining module 1102, configured to obtain a ground point cloud to be detected;
the grid map building module 1104 is used for building a grid map according to the ground point clouds and determining two-dimensional ground points covered by each grid in the grid map;
and the detection analysis module 1106 is configured to obtain a height value of the two-dimensional ground point covered by each grid, and determine the flatness of the road surface to be detected according to the height value of the two-dimensional ground point covered by each grid.
The ground point cloud obtaining module 1102 includes:
the acquisition unit is used for acquiring point cloud data of the acquired road surface to be detected;
and the extraction unit is used for extracting the ground point cloud of the road surface to be detected from the point cloud data.
The extraction unit is also used for acquiring a preset number of point clouds from the point cloud data from near to far according to a distance acquisition device, determining ground point clouds in the preset number of point clouds according to the height of the acquisition device, and performing straight line fitting according to the determined ground point clouds to determine the ground point clouds in the point cloud data.
Wherein, the grid map building module 1104 includes:
the projection unit is used for converting the ground point cloud to be detected into two-dimensional ground points according to the three-dimensional information of the ground point cloud to be detected;
and the grid dividing unit is used for covering the two-dimensional ground points by adopting a grid map with the grid size of a first preset value to obtain a grid map.
Wherein, the detection analysis module 1106 includes:
the data acquisition unit is used for acquiring the height value of the two-dimensional ground point covered by each grid;
the statistical unit is used for carrying out statistics on the height value of the two-dimensional ground point covered by each grid to obtain the statistical value of each grid;
the first calculation unit is used for constructing a template with the side length as a second preset value by taking each grid as a center, and determining the flatness of each grid according to the statistic value of each grid in the template;
and the second calculating unit is used for acquiring the average value or the median value of the flatness of each grid, and taking the average value or the median value of the flatness of each grid as the flatness of the road surface to be detected.
The statistical unit is further configured to traverse the two-dimensional ground points covered by each grid to obtain a maximum height value, a height value average value, or a minimum height value of the two-dimensional ground points covered by each grid, and use the maximum height value, the height value average value, or the minimum height value of the two-dimensional ground points covered by each grid as a corresponding statistical value of each grid.
The detection analysis module 1106 further includes: and the interpolation processing unit is used for carrying out interpolation processing on the grids with the empty statistical values.
The interpolation processing unit is also used for constructing a template with the side length being a third preset value by taking the grid with the empty statistical value as a center; and acquiring the mean value of the statistical value of each grid in the constructed template, and taking the mean value of the statistical value as the interpolation of the grid with the empty statistical value.
The road surface detection device may further include:
the pit detection module is used for carrying out pit detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid;
and the fault detection module is used for carrying out fault detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid.
For specific limitations of the road surface detection device, reference may be made to the above limitations of the road surface detection method, which are not described herein again. The modules in the road surface detection device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 12. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a road surface detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 12 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a computer apparatus including a memory in which a computer program is stored and a processor that implements the above-described road surface detecting step when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the above-described road surface detecting step.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (9)
1. A method of road surface inspection, the method comprising:
acquiring point cloud data of a road surface to be detected, which is acquired by a laser radar;
acquiring a preset number of point clouds from the point cloud data according to the distance from the laser radar to the laser radar;
taking a vertical distance between a contact point of a tire of the laser radar and the ground and the laser radar as a height reference value, and if a difference value between the height reference value and a height value of a point cloud is within a preset interval, taking the point cloud as a ground point cloud until two point clouds are the ground point cloud, namely a first ground point cloud and a second ground point cloud respectively;
performing linear fitting and phase extension processing according to the first ground point cloud and the second ground point cloud to determine ground point cloud in the point cloud data;
constructing a grid map according to the ground point cloud, and determining two-dimensional ground points covered by each grid in the grid map;
traversing the two-dimensional ground points covered by each grid to obtain the maximum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid;
taking the maximum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid as the corresponding statistical value of each grid;
for each grid, taking a certain grid as a central grid, calculating the variance of the statistical values of the grids in the predetermined range around the central grid according to the statistical values of the grids in the predetermined range around the central grid, and determining the flatness of the central grid according to the variance;
and acquiring the average value or the median value of the flatness of each grid, and taking the average value or the median value of the flatness of each grid as the flatness of the pavement to be detected.
2. The method of claim 1, wherein the constructing a grid map from the ground point clouds and determining the two-dimensional ground points covered by each grid in the grid map comprises:
converting the ground point cloud to be detected into two-dimensional ground points according to the three-dimensional information of the ground point cloud to be detected;
and covering the two-dimensional ground points by adopting a grid map with the grid size of a first preset value to obtain a grid map.
3. The method of claim 1, wherein after statistically analyzing the height values of the two-dimensional ground points covered by each grid to obtain statistical values of each grid, the method further comprises: and carrying out interpolation processing on the grids with empty statistical values.
4. The road surface detection method according to claim 3, wherein the interpolation processing of the grid whose statistical value is empty includes:
constructing a template with the side length of a third preset value by taking the grid with the empty statistical value as a center;
and acquiring the mean value of the statistical value of each grid in the constructed template, and taking the mean value of the statistical value as the interpolation of the grid with the empty statistical value.
5. The road surface detection method according to claim 1, characterized by further comprising:
and carrying out pit detection or fault detection on the ground to be detected according to the height value of the two-dimensional ground point covered by each grid.
6. A road surface detecting device, characterized in that the device comprises:
the ground point cloud acquisition module is used for acquiring point cloud data of a to-be-detected road surface acquired by a laser radar; acquiring a preset number of point clouds from the point cloud data according to the distance from the laser radar to the laser radar; taking a vertical distance between a contact point of a tire of the laser radar and the ground and the laser radar as a height reference value, and if a difference value between the height reference value and a height value of a point cloud is within a preset interval, taking the point cloud as a ground point cloud until two point clouds are the ground point cloud, namely a first ground point cloud and a second ground point cloud respectively; performing linear fitting and phase extension processing according to the first ground point cloud and the second ground point cloud to determine ground point cloud in the point cloud data;
the grid map building module is used for building a grid map according to the ground point clouds and determining two-dimensional ground points covered by each grid in the grid map;
the detection analysis module is used for traversing the two-dimensional ground points covered by each grid to obtain the maximum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid; taking the maximum height value or the average height value or the minimum height value of the two-dimensional ground points covered by each grid as the corresponding statistical value of each grid; for each grid, taking a certain grid as a central grid, calculating the variance of the statistical values of the grids in the predetermined range around the central grid according to the statistical values of the grids in the predetermined range around the central grid, and determining the flatness of the central grid according to the variance; and acquiring the average value or the median value of the flatness of each grid, and taking the average value or the median value of the flatness of each grid as the flatness of the pavement to be detected.
7. The apparatus of claim 6, wherein the grid map construction module comprises:
the projection unit is used for converting the ground point cloud to be detected into two-dimensional ground points according to the three-dimensional information of the ground point cloud to be detected;
and the grid dividing unit is used for covering the two-dimensional ground points by adopting a grid map with the grid size of a first preset value to obtain a grid map.
8. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 5 when executing the computer program.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910220802.6A CN110084116B (en) | 2019-03-22 | 2019-03-22 | Road surface detection method, road surface detection device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910220802.6A CN110084116B (en) | 2019-03-22 | 2019-03-22 | Road surface detection method, road surface detection device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110084116A CN110084116A (en) | 2019-08-02 |
| CN110084116B true CN110084116B (en) | 2022-02-01 |
Family
ID=67413442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910220802.6A Active CN110084116B (en) | 2019-03-22 | 2019-03-22 | Road surface detection method, road surface detection device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084116B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110487245A (en) * | 2019-08-16 | 2019-11-22 | 中国民用航空总局第二研究所 | Civil airport glide-path localizer protection zone place real-time monitoring analyzing system and its method |
| CA3191715A1 (en) * | 2019-09-22 | 2021-03-25 | Vayavision Sensing Ltd. | Functional safety in autonomous driving |
| CN111021206B (en) * | 2019-11-20 | 2021-06-25 | 中铁四局集团第一工程有限公司 | Road surface flatness detection method and system |
| CN111123278B (en) * | 2019-12-30 | 2022-07-12 | 科沃斯机器人股份有限公司 | Partitioning method, partitioning equipment and storage medium |
| CN113129255B (en) * | 2019-12-31 | 2023-04-07 | 杭州海康机器人股份有限公司 | Method, computing device, system and storage medium for detecting package |
| CN111208533A (en) * | 2020-01-09 | 2020-05-29 | 上海工程技术大学 | Real-time ground detection method based on laser radar |
| CN111208530B (en) * | 2020-01-15 | 2022-06-17 | 北京四维图新科技股份有限公司 | Positioning layer generation method and device, high-precision map and high-precision map equipment |
| CN111999741B (en) * | 2020-01-17 | 2023-03-14 | 青岛慧拓智能机器有限公司 | Method and device for detecting roadside laser radar target |
| CN111239757B (en) * | 2020-03-12 | 2022-04-19 | 湖南大学 | Automatic extraction method and system for road surface characteristic parameters |
| CN113748693B (en) * | 2020-03-27 | 2023-09-15 | 深圳市速腾聚创科技有限公司 | Position and pose correction method and device of roadbed sensor and roadbed sensor |
| CN112199991B (en) * | 2020-08-27 | 2024-04-30 | 广州中国科学院软件应用技术研究所 | Simulation point cloud filtering method and system applied to vehicle-road cooperation road side perception |
| CN112129266B (en) * | 2020-09-28 | 2022-12-16 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, device and computer readable storage medium for processing map |
| CN111932943B (en) * | 2020-10-15 | 2021-05-14 | 深圳市速腾聚创科技有限公司 | Dynamic target detection method and device, storage medium and roadbed monitoring equipment |
| CN112348781A (en) * | 2020-10-26 | 2021-02-09 | 广东博智林机器人有限公司 | Method, device and equipment for detecting height of reference plane and storage medium |
| CN113970295B (en) * | 2021-09-28 | 2024-04-16 | 湖南三一中益机械有限公司 | Spreading thickness measuring method and device and spreading machine |
| CN114322856B (en) * | 2021-12-16 | 2023-09-15 | 青岛慧拓智能机器有限公司 | Mining area pavement evenness detection method, device, storage medium and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127771A (en) * | 2016-06-28 | 2016-11-16 | 上海数联空间科技有限公司 | Tunnel orthography system and method is obtained based on laser radar LIDAR cloud data |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106650640A (en) * | 2016-12-05 | 2017-05-10 | 浙江大学 | Negative obstacle detection method based on local structure feature of laser radar point cloud |
| CN107092020A (en) * | 2017-04-19 | 2017-08-25 | 北京大学 | Merge the surface evenness monitoring method of unmanned plane LiDAR and high score image |
-
2019
- 2019-03-22 CN CN201910220802.6A patent/CN110084116B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127771A (en) * | 2016-06-28 | 2016-11-16 | 上海数联空间科技有限公司 | Tunnel orthography system and method is obtained based on laser radar LIDAR cloud data |
| CN106199558A (en) * | 2016-08-18 | 2016-12-07 | 宁波傲视智绘光电科技有限公司 | Barrier method for quick |
| CN106650640A (en) * | 2016-12-05 | 2017-05-10 | 浙江大学 | Negative obstacle detection method based on local structure feature of laser radar point cloud |
| CN107092020A (en) * | 2017-04-19 | 2017-08-25 | 北京大学 | Merge the surface evenness monitoring method of unmanned plane LiDAR and high score image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110084116A (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084116B (en) | Road surface detection method, road surface detection device, computer equipment and storage medium | |
| CN111160302A (en) | Obstacle information identification method and device based on automatic driving environment | |
| US20160133008A1 (en) | Crack data collection method and crack data collection program | |
| CN111144228B (en) | Obstacle identification method based on 3D point cloud data and computer equipment | |
| US20160133007A1 (en) | Crack data collection apparatus and server apparatus to collect crack data | |
| WO2009143986A1 (en) | Automated building outline detection | |
| CN111292275B (en) | Point cloud data filtering method and device based on complex ground and computer equipment | |
| CN112149572A (en) | Road edge detection method, device and storage medium | |
| CN111243003A (en) | Vehicle-mounted binocular camera and method and device for detecting road height limiting rod | |
| CN114004950B (en) | BIM and LiDAR technology-based intelligent pavement disease identification and management method | |
| Haghighatgou et al. | A method for automatic identification of openings in buildings facades based on mobile LiDAR point clouds for assessing impacts of floodings | |
| CN114663598A (en) | Three-dimensional modeling method, device and storage medium | |
| CN114842166A (en) | Negative obstacle detection method, system, medium, and apparatus applied to structured road | |
| KR102260556B1 (en) | Deep learning-based parking slot detection method and apparatus integrating global and local information | |
| CN116958099B (en) | Cable abrasion detection method, system, device and computer equipment | |
| CN110174115B (en) | Method and device for automatically generating high-precision positioning map based on perception data | |
| Yan et al. | Quality analysis on ransac-based roof facets extraction from airborne lidar data | |
| JP2019046295A (en) | Monitoring device | |
| CN111862206A (en) | Visual positioning method and device, electronic equipment and readable storage medium | |
| CN116894885A (en) | Map data generation method and apparatus, and computer-readable recording medium storing map data generation program | |
| CN111160266B (en) | Object tracking method and device | |
| CN113393423A (en) | Cliff detection method and device based on point cloud and mobile robot | |
| CN113050063A (en) | Obstacle detection method, device and equipment based on laser sensor and storage medium | |
| CN117315001B (en) | Method and system for rapidly detecting building area based on laser scanning | |
| Jang et al. | Deep learning-based 3D digital damage map of vertical-type tunnels using unmanned fusion data scanning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |