CN109359614B - Laser point cloud plane identification method, device, equipment and medium - Google Patents
Laser point cloud plane identification method, device, equipment and medium Download PDFInfo
- Publication number
- CN109359614B CN109359614B CN201811272869.6A CN201811272869A CN109359614B CN 109359614 B CN109359614 B CN 109359614B CN 201811272869 A CN201811272869 A CN 201811272869A CN 109359614 B CN109359614 B CN 109359614B
- Authority
- CN
- China
- Prior art keywords
- plane
- grid
- point cloud
- cloud data
- laser 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Electromagnetism (AREA)
- Traffic Control Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The embodiment of the invention discloses a plane identification method, a plane identification device, plane identification equipment and a plane identification medium for laser point clouds. The method comprises the following steps: acquiring a frame of laser point cloud data acquired by a vehicle acquisition device; performing raster division on the laser point cloud data; performing plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center; and processing according to the plane recognition result. The embodiment of the invention solves the problem that the accuracy of the identification result is low because the differential treatment of the plane identification is not carried out according to the characteristics of the laser point cloud data, realizes the dynamic setting of the threshold value in the plane identification according to the characteristics of the laser point cloud, and can improve the accuracy of the plane identification.
Description
Technical Field
The embodiment of the invention relates to a data processing technology, in particular to a plane identification method, a device, equipment and a medium for laser point cloud.
Background
In the technical fields of automatic driving and assisted driving, environmental sensing using a lidar is a widely used technique.
In the prior art, a laser ray is transmitted through a laser radar, and the laser ray is reflected back when encountering an impenetrable obstacle and is received by receiving equipment of the laser radar; and calculating the distance between the obstacle and the laser radar by recording the time delay between the emission and the reception and combining the transmission speed of the ray. By detecting a large amount of laser beams, the sensing of the surrounding environment can be performed. Taking 64 line lidar as an example, 130 million points per second may be generated, each point reflecting a corresponding distance. The point cloud data obtained by one detection of the laser radar can be called as a frame of laser point cloud data.
After the laser point cloud data is obtained, data processing is required to be further performed, and various obstacles such as a cart, a trolley, pedestrians and the like are identified based on the laser point cloud data, so that the laser point cloud data is used as a basis for controlling the running of the vehicle. The plane recognition is the basic processing operation before the obstacle recognition, but the prior art does not combine the characteristics of the laser point cloud to carry out the differential treatment of the plane recognition, so that the accuracy of the recognition result is lower.
Disclosure of Invention
The embodiment of the invention provides a plane identification method, a plane identification device, plane identification equipment and a plane identification medium of laser point cloud, which are used for carrying out plane identification by combining the characteristics of the laser point cloud and improving the identification accuracy.
In a first aspect, an embodiment of the present invention provides a plane identification method for a laser point cloud, where the method includes:
acquiring a frame of laser point cloud data acquired by a vehicle acquisition device;
performing raster division on the laser point cloud data;
performing plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center;
and processing according to the plane recognition result.
In a second aspect, an embodiment of the present invention further provides a device for identifying a plane of a laser point cloud, where the device includes:
the point cloud data acquisition module is used for acquiring a frame of laser point cloud data acquired by the vehicle acquisition device;
the grid division module is used for carrying out grid division on the laser point cloud data;
the plane identification module is used for carrying out plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center;
and the processing module is used for processing according to the plane recognition result.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
storage means for storing one or more programs;
when the one or more programs are executed by the one or more processors, the one or more processors implement the method for identifying a plane of a laser point cloud according to any of the embodiments of the present invention.
In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the plane identification method for a laser point cloud according to any one of the embodiments of the present invention.
According to the embodiment of the invention, laser point cloud data are obtained and subjected to raster division, and then during plane identification, a plane identification algorithm and a set point number threshold are adopted for each raster, wherein the set point number threshold adopted by each raster is different, and then processing is carried out according to a plane identification result; the problem that plane identification is not differentially treated according to the characteristics of laser point cloud data, so that the accuracy of an identification result is low is solved, the threshold is dynamically set in the plane identification according to the characteristics of the laser point cloud, and the accuracy of the plane identification can be improved.
Drawings
FIG. 1 is a flowchart of a method for identifying a plane of a laser point cloud according to a first embodiment of the present invention;
FIG. 2a is a flowchart of a method for identifying a plane of a laser point cloud according to a second embodiment of the present invention;
FIG. 2b is a diagram illustrating the effect of filtering the ground in the laser point cloud data according to the second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a plane recognition device for laser point cloud according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device in the fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a plane recognition method for laser point cloud according to an embodiment of the present invention, which is applicable to a situation of plane recognition through laser point cloud data, and the method may be implemented by a plane recognition device for laser point cloud, and may be implemented by software and/or hardware in a device, and the device may be integrated into any device for processing laser point cloud data, and optionally, may be a vehicle-mounted control device. As shown in fig. 1, the method for identifying a plane by laser point cloud specifically includes:
and S110, acquiring a frame of laser point cloud data acquired by the vehicle acquisition unit.
The laser point cloud data may be laser point cloud data collected on a road to be measured by a laser scanning device carried by a vehicle (such as an unmanned automobile). Laser scanning devices are typically placed outside the vehicle body, such as on the roof of a vehicle. The point cloud data obtained by one-time detection of the laser scanning device may be called a frame of laser point cloud data, and one frame of laser point cloud data may be selected optionally for analysis and identification of the point cloud data.
And S120, performing raster division on the laser point cloud data.
Road surface points scanned by a scanning device such as a laser radar are not completely flat and are discontinuous, and therefore, the laser point cloud data is divided into grids, and the size of each grid is the same. In one embodiment, the grid may be provided as a cube having a length of 8 meters, depending on the distance that the laser scanning apparatus can scan. Due to the characteristics of the laser point cloud, point cloud data points from near to far become sparse gradually by taking a vehicle as a center. That is, the closer to the laser light source, the larger the number of point clouds.
And S130, carrying out plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center.
Specifically, for each grid, a plane recognition algorithm may be selected for plane recognition, and in the recognition process, a threshold value of the set point number is usually adopted as a limit value for recognizing the plane. Because the sparsity of the data points in each grid is originally different, the set point number threshold values adopted by setting each grid are different and are changed according to the distance between the grid and the acquisition center, so that the sparse situation of the data points in each grid is adapted, and the accuracy of plane identification is higher.
The RANSAC algorithm is used as an example for explanation. And performing plane iterative identification by using a RANSAC algorithm until the number of the remaining data points which are not belonged to the plane in the grid is less than a set point number threshold value, and stopping plane identification.
The random Sample consensus (random Sample consensus) random sampling consistency method is to estimate mathematical model parameters (the model is a spatial plane model herein) in an iterative manner from a set of samples containing "outliers". The specific process comprises the steps of firstly providing a mathematical model to be estimated, then randomly selecting a sample subset from observation data samples, calculating model parameters of the sample subset through a minimum variance estimation method, then calculating the variance between all the observation data samples and the model, counting sample points (inliers) in the model and local points (outliers) outside the model in the observation data according to preset threshold values and variances, and repeating the process according to iteration times. And at the end of each iteration, calculating a factor for judging the end of the current iteration according to the current iteration times, the model parameters, the expected error rate and the total number of samples, and determining whether to end the iteration process. At the end of the iterative process, the model parameters are the optimal estimated parameter values.
In this embodiment, the ending factor of the current iteration is the set threshold of the number of set points for different grids.
And S140, processing according to the plane recognition result.
After the plane recognition is performed, a plurality of planes can be obtained, wherein the planes comprise building surfaces, vehicle surfaces, lawns, road surfaces and the like. The road surface data occupies approximately one fourth of the total data amount, and the data processing load is increased in obstacle recognition. Therefore, based on the subsequent requirement of identifying the obstacles and the like, the point cloud data corresponding to the road surface can be filtered from the laser point cloud data on the basis of plane identification.
According to the technical scheme, laser point cloud data are obtained and subjected to grid division, so that a RANSAC random sampling consistency plane identification algorithm is adopted for each grid during plane identification, and different set point threshold values are set for different grids to serve as iteration end factors; the problem that the accuracy of the recognition result is low due to the fact that differential treatment of plane recognition is not carried out according to the characteristics of the laser point cloud data is solved, the threshold is dynamically set in the plane recognition according to the characteristics of the laser point cloud data, the accuracy of the plane recognition and the robustness of a road surface segmentation method can be improved, and the processing efficiency of subsequent obstacle recognition can be improved after the road surface is segmented.
Example two
Fig. 2a is a flowchart of a plane identification method for laser point cloud according to a second embodiment of the present invention. Based on the alternatives in the above embodiments, the present embodiment provides a process of filtering point cloud data corresponding to a road surface from laser point cloud data according to a plane recognition result. As shown in fig. 2a, the method for identifying a plane of a laser point cloud provided in the embodiment of the present invention includes the following steps:
s210, acquiring a frame of laser point cloud data acquired by the vehicle acquisition unit.
And S220, performing raster division on the laser point cloud data.
And S230, acquiring the number of data points in each grid, and multiplying the number by a set proportion to be used as a set point number threshold of the grid.
Due to the characteristics of the laser point cloud, point cloud points from near to far gradually become sparse by taking a vehicle collector as a center, and therefore, a set point number threshold value set for carrying out plane identification on each grid can be determined according to the grid position. When the number of points not belonging to any plane in a grid is less than or equal to a set point threshold value, the identification of the plane in the grid can be finished.
In one embodiment, assuming that the number of point cloud points within the grid is N (N is a positive integer), 10% of N may be taken as the set point number threshold for the grid. Wherein, 10% is a set proportion, and the set proportion can be set according to the actual laser point cloud data volume, so that the set point threshold of each grid and the point cloud number in the grid form a certain proportional relation.
In another embodiment, the threshold number of set points for each grid may be determined based on setting an inverse relationship based on the distance between the grid and the acquisition center. For example, when the grid is 10 meters from the acquisition center, the set point number threshold is 100; when the distance between the grid and the acquisition center is 20 meters, the set point number threshold is 90; the set point number threshold is 80 when the grid is 20 meters from the acquisition center. Wherein, the collection center is the position of the vehicle collector.
And S240, performing plane identification on each grid by adopting a plane identification algorithm and a set point number threshold, wherein the set point number threshold adopted by each grid is different according to the distance difference between the grid and the acquisition center.
Specifically, a RANSAC method is used for extracting a plane according to a dynamic threshold value of each grid, and when the number of remaining points of an iterative extraction plane is smaller than the dynamic threshold value, the plane extraction is stopped. For example, when the number of points remaining after extracting the plane is less than N × 10%, the iteration is stopped and the extracting the plane is ended.
And S250, filtering point cloud data corresponding to the road surface from the laser point cloud data according to the plane identification result.
And determining a plane belonging to the road surface in the identified plane, and judging the plane through an included angle and a height between a normal vector of the plane and the vertical direction.
Specifically, the road surface is usually a horizontal plane or an inclined plane with a certain angle, the included angle between the normal vector of the road surface and the vertical direction is equal to or close to 90 degrees, and if the included angle between the normal vector of any plane and the vertical direction is smaller than a set included angle threshold value, the plane is determined to be a non-road surface.
Generally, a laser scanning device is placed at the top of a laser point cloud data acquisition vehicle, the laser scanning device is a data acquisition center, namely the vertical distance between the data acquisition center and a road surface is the height of the laser point cloud data acquisition vehicle, if any plane is not matched with a height threshold value, the plane is determined to be a non-road surface, wherein the height threshold value is set according to the height of the vehicle and the average height of point cloud data points in a grid.
And then, determining a target road surface by combining road surface identification results of the included angle threshold and the height threshold, and filtering point cloud data corresponding to the target road surface from the laser point cloud data. It should be noted that there is no fixed order between the above two aspects.
Further, if any plane does not match the height threshold, determining that the plane is a non-road surface comprises:
according to the point cloud data in each grid, counting the average height value of each data point in the grid; determining the dynamic height threshold of each grid according to the known vehicle height value and the average height value, specifically, calculating a proportional reduced value of the known vehicle height value and a proportional reduced value of the average height value, and taking the maximum value of the proportional reduced value and the average height value as the dynamic height threshold of the grid, wherein the reduced ratio of the known vehicle height value and the reduced ratio of the average height value can be customized; and calculating a plane distance value from the center point of each grid to each plane, and if the plane distance value is greater than the dynamic height threshold value, determining that the plane is a non-road surface.
In one embodiment, the process of screening planes according to their height is as follows: first, from the point cloud data in each grid, the average height feature Z of the points in the grid is counted, the vehicle height H is known, and then the dynamic height threshold of the plane can be set to max (Z/2, H/5), where 2 is the reduced ratio of the average height value and 5 is the reduced ratio of the vehicle height value. And calculating the distance characteristic D from the central point of the grid to the extracted space plane for each plane, wherein if the D is larger than the dynamic height threshold value, the extracted space plane point is not a ground point. Points excluding the roof plane and the green belt roof plane are recognized as ground points.
In this embodiment, the effect of filtering the ground from the laser point cloud data can refer to fig. 2b, in fig. 2b, white points or a set of points are point cloud points of non-road points, after deleting the road point cloud points, two boundary lines between the obstacle point cloud and the road surface can be clearly seen, the middle area of the two boundary lines is the road surface, obstacle point cloud points with a smaller volume are further arranged on the road surface, the obstacle point cloud points are more concentrated on two sides of the road surface, and the obstacles can be buildings, numbers and the like.
According to the technical scheme of the embodiment, the number of data points in each grid is obtained by obtaining laser point cloud data and carrying out grid division on the laser point cloud data, the number is multiplied by a set proportion to serve as a set point number threshold of the grid, or the set point number threshold of each grid can be determined based on a set inverse proportion relation according to the distance between each grid and an acquisition center, then when plane identification is carried out, plane identification is carried out on each grid by adopting a plane identification algorithm and the set point number threshold, and then point cloud data corresponding to a road surface are filtered out from the laser point cloud data; the problem that the accuracy of the recognition result is low due to the fact that differential treatment of plane recognition is not carried out according to the characteristics of the laser point cloud data is solved, the threshold is dynamically set in the plane recognition according to the characteristics of the laser point cloud data, the accuracy of the plane recognition and the robustness of a road surface segmentation method can be improved, and the processing efficiency of subsequent obstacle recognition can be improved after the road surface is segmented.
EXAMPLE III
Fig. 3 shows a schematic structural diagram of a plane recognition device for laser point cloud provided in the third embodiment of the present disclosure, which is applicable to the case of performing plane recognition based on characteristics of laser point cloud data.
As shown in fig. 3, the plane recognition apparatus for laser point cloud in the embodiment of the present disclosure includes: a point cloud data acquisition module 310, a raster division module 320, a plane identification module 330, and a processing module 340.
The point cloud data acquisition module 310 is used for acquiring a frame of laser point cloud data acquired by the vehicle acquisition unit; a grid division module 320, configured to perform grid division on the laser point cloud data; a plane identification module 330, configured to perform plane identification by using a plane identification algorithm and a set point number threshold for each grid, where the set point number threshold used for each grid is different according to a distance difference between the grid and the acquisition center; and the processing module 340 is configured to perform processing according to the plane recognition result.
According to the technical scheme of the embodiment, laser point cloud data are obtained and subjected to raster division, and then during plane identification, plane identification is carried out on each raster by adopting a plane identification algorithm and a set point number threshold, wherein the set point number thresholds adopted by each raster are different, and then processing is carried out according to a plane identification result; the problem that the accuracy of the recognition result is low due to the fact that differential treatment of plane recognition is not carried out according to the characteristics of the laser point cloud data is solved, the threshold is dynamically set in the plane recognition according to the characteristics of the laser point cloud, the accuracy of the plane recognition and the robustness of a road surface segmentation method can be improved, and the processing efficiency of subsequent obstacle recognition can be improved after the road surface is segmented.
Further, the processing module 340 is specifically configured to:
and filtering point cloud data corresponding to the road surface from the laser point cloud data according to the plane identification result.
Optionally, the plane identification module 330 is specifically configured to:
and performing plane iterative identification on each grid by adopting a RANSAC algorithm until the number of the remaining data points which do not belong to the plane in the grid is less than the set point threshold, and stopping the plane identification.
Further, the laser point cloud plane recognition device further comprises a point number threshold setting module, which is used for acquiring the number of data points in each grid before plane recognition is carried out on each grid by adopting a plane recognition algorithm and a set point number threshold, and multiplying the number by a set proportion to be used as the set point number threshold of the grid; or
And determining a set point number threshold value of each grid based on a set inverse relation according to the distance between each grid and the acquisition center.
Optionally, the processing module 340 includes a first non-road surface recognition unit, a second non-road surface recognition unit and a processing unit; wherein the content of the first and second substances,
the first non-road surface identification unit is used for determining that any plane is a non-road surface when an included angle between a normal vector of the plane and the vertical direction is smaller than a set included angle threshold value;
the second non-road surface identification unit is used for determining that any plane is a non-road surface when the plane is not matched with a height threshold value, wherein the height threshold value is set according to the height of the vehicle and the average height of the point cloud data points in the grid;
and the processing unit is used for determining the target road surface by combining the road surface identification results of the included angle threshold and the height threshold, and filtering point cloud data corresponding to the target road surface from the laser point cloud data.
Further, the second non-road surface recognition unit includes: the system comprises an average height statistics subunit, a dynamic height threshold determination subunit and a non-road surface identification subunit; wherein the content of the first and second substances,
the average height counting subunit is used for counting the average height value of each data point in the grids according to the point cloud data in each grid;
a dynamic height threshold determination subunit, configured to determine a dynamic height threshold for each grid according to a known vehicle height value and the average height value;
and the non-road surface identification subunit is used for calculating a plane distance value from the center point of each grid to each plane, and if the plane distance value is greater than the dynamic height threshold value, determining that the plane is a non-road surface.
Further, the dynamic height threshold determining subunit is specifically configured to calculate a proportional reduced value of the known vehicle height value and a proportional reduced value of the average height value, and take a maximum value of the proportional reduced value and the average height value as the dynamic height threshold of the grid.
The plane recognition device for the laser point cloud provided by the embodiment of the invention can execute the plane recognition method for the laser point cloud provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 4 is a schematic structural diagram of a computer device in the fourth embodiment of the present invention. FIG. 4 illustrates a block diagram of an exemplary computer device 412 suitable for use in implementing embodiments of the present invention. The computer device 412 shown in FIG. 4 is only one example and should not impose any limitations on the functionality or scope of use of embodiments of the present invention. The computer device 412 is preferably an in-vehicle control device.
As shown in FIG. 4, computer device 412 is in the form of a general purpose computing device. Components of computer device 412 may include, but are not limited to: one or more processors or processing units 416, a system memory 428, and a bus 418 that couples the various system components including the system memory 428 and the processing unit 416.
The system memory 428 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)430 and/or cache memory 432. The computer device 412 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 434 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 4, commonly referred to as a "hard drive"). Although not shown in FIG. 4, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 418 by one or more data media interfaces. Memory 428 can include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 440 having a set (at least one) of program modules 442 may be stored, for instance, in memory 428, such program modules 442 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 442 generally perform the functions and/or methodologies of the described embodiments of the invention.
The computer device 412 may also communicate with one or more external devices 414 (e.g., keyboard, pointing device, display 424, etc.), with one or more devices that enable a user to interact with the computer device 412, and/or with any devices (e.g., network card, modem, etc.) that enable the computer device 412 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 422. Also, computer device 412 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet) through network adapter 420. As shown, network adapter 420 communicates with the other modules of computer device 412 over bus 418. It should be appreciated that although not shown in FIG. 4, other hardware and/or software modules may be used in conjunction with the computer device 412, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 416 executes programs stored in the system memory 428 to execute various functional applications and data processing, for example, implement the plane identification method for laser point cloud provided by the embodiment of the present invention, the method mainly includes:
acquiring a frame of laser point cloud data acquired by a vehicle acquisition device;
performing raster division on the laser point cloud data;
performing plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center;
and processing according to the plane recognition result.
EXAMPLE five
The fifth embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the plane identification method for laser point cloud provided in the fifth embodiment of the present invention, where the method mainly includes:
acquiring a frame of laser point cloud data acquired by a vehicle acquisition device;
performing raster division on the laser point cloud data;
performing plane identification by adopting a plane identification algorithm and a set point number threshold value aiming at each grid, wherein the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the acquisition center;
and processing according to the plane recognition result.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (9)
1. A plane identification method of laser point cloud is characterized by comprising the following steps:
acquiring a frame of laser point cloud data acquired by a vehicle acquisition device;
performing raster division on the laser point cloud data;
performing plane iterative identification on each grid by adopting a RANSAC algorithm through parameters of an iterative space plane model until the number of remaining data points which do not belong to a plane in the grid is less than a set point number threshold, and stopping the plane identification, wherein the set point number threshold adopted by each grid is different according to the distance difference between the grid and a vehicle collector; the set point number threshold for the grid is determined by: determining a set point number threshold value of each grid based on a set inverse relation according to the distance between the grid and a vehicle collector;
and processing according to the plane recognition result.
2. The method of claim 1, wherein processing based on the plane recognition result comprises:
and filtering point cloud data corresponding to the road surface from the laser point cloud data according to the plane identification result.
3. The method of any of claims 1-2, wherein the threshold number of set points for the grid is further determined by:
and acquiring the number of data points in the grid, and multiplying the number by a set proportion to be used as a set point number threshold value of the grid.
4. The method of claim 2, wherein filtering point cloud data corresponding to the road surface from the laser point cloud data according to the plane identification result comprises:
if the included angle between the normal vector of any plane and the vertical direction is smaller than a set included angle threshold value, determining that the plane is a non-road surface;
if any plane is not matched with a height threshold value, determining that the plane is a non-road surface, wherein the height threshold value is set according to the height of the vehicle and the average height of the point cloud data points in the grid;
and determining a target road surface by combining the road surface identification results of the included angle threshold and the height threshold, and filtering point cloud data corresponding to the target road surface from the laser point cloud data.
5. The method of claim 4, wherein if any plane does not match the height threshold, determining that the plane is off-road comprises:
according to the point cloud data in each grid, counting the average height value of each data point in the grid;
determining a dynamic height threshold for each grid from the known vehicle height values and the average height value;
and calculating a plane distance value from the center point of each grid to each plane, and if the plane distance value is greater than the dynamic height threshold value, determining that the plane is a non-road surface.
6. The method of claim 5, wherein determining a dynamic height threshold for each grid based on a known vehicle height value and the average height value comprises:
and calculating a proportional reduced value of the height value of the known vehicle and a proportional reduced value of the average height value, and taking the maximum value of the proportional reduced value and the average height value as the dynamic height threshold value of the grid.
7. A plane recognition device of laser point cloud is characterized by comprising:
the point cloud data acquisition module is used for acquiring a frame of laser point cloud data acquired by the vehicle acquisition device;
the grid division module is used for carrying out grid division on the laser point cloud data;
the plane identification module is used for carrying out plane iterative identification on each grid by adopting a RANSAC algorithm through parameters of an iterative space plane model until the number of the remaining data points which do not belong to the plane in the grid is less than a set point number threshold value, and the set point number threshold value adopted by each grid is different according to the distance difference between the grid and the vehicle collector; the set point number threshold for the grid is determined by: determining a set point number threshold value of each grid based on a set inverse relation according to the distance between the grid and a vehicle collector;
and the processing module is used for processing according to the plane recognition result.
8. A computer device, characterized in that the computer device comprises:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of plane identification of a laser point cloud of any of claims 1-6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a method of plane recognition of a laser point cloud according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811272869.6A CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811272869.6A CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109359614A CN109359614A (en) | 2019-02-19 |
| CN109359614B true CN109359614B (en) | 2021-06-11 |
Family
ID=65347367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811272869.6A Active CN109359614B (en) | 2018-10-30 | 2018-10-30 | Laser point cloud plane identification method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109359614B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458805B (en) * | 2019-03-26 | 2022-05-13 | 华为技术有限公司 | Plane detection method, computing device and circuit system |
| JP7165630B2 (en) * | 2019-07-24 | 2022-11-04 | 本田技研工業株式会社 | Recognition system, vehicle control system, recognition method, and program |
| CN110516564A (en) * | 2019-08-06 | 2019-11-29 | 深兰科技（上海）有限公司 | Pavement detection method and apparatus |
| CN110471074B (en) * | 2019-08-15 | 2021-04-27 | 岭纬科技（厦门）有限公司 | Method for improving planar laser point cloud distance measurement precision |
| CN110782472B (en) * | 2019-09-05 | 2022-06-07 | 腾讯科技（深圳）有限公司 | Point cloud ground point identification method and device |
| CN112634181B (en) * | 2019-09-24 | 2024-06-14 | 阿波罗智能技术(北京)有限公司 | Method and device for detecting ground point cloud points |
| CN112634487B (en) * | 2019-09-24 | 2022-08-16 | 北京百度网讯科技有限公司 | Method and apparatus for outputting information |
| CN111027601B (en) * | 2019-11-25 | 2023-10-17 | 歌尔股份有限公司 | Plane detection method and device based on laser sensor |
| CN111611900B (en) * | 2020-05-15 | 2023-06-30 | 北京京东乾石科技有限公司 | Target point cloud identification method and device, electronic equipment and storage medium |
| CN113219446A (en) * | 2021-04-30 | 2021-08-06 | 森思泰克河北科技有限公司 | In-vehicle radar occupancy identification method and device and vehicle-mounted radar |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
| CN106408604A (en) * | 2016-09-22 | 2017-02-15 | 北京数字绿土科技有限公司 | Filtering method and device for point cloud data |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855663B (en) * | 2012-05-04 | 2015-04-01 | 北京建筑工程学院 | Method for building CSG (Constructive Solid Geometry) model according to laser radar grid point cloud |
| CN104251662B (en) * | 2013-06-27 | 2017-10-31 | 杭州中科天维科技有限公司 | Cloud threshold adaptive noise suppression technology is put in order |
| CN104156926B (en) * | 2014-08-19 | 2017-06-23 | 武汉海达数云技术有限公司 | Vehicle-mounted laser point cloud noise spot minimizing technology under many scenes |
| CN104143194B (en) * | 2014-08-20 | 2017-09-08 | 清华大学 | A kind of point cloud segmentation method and device |
| CN104715482B (en) * | 2015-03-20 | 2017-08-11 | 四川大学 | A kind of RANSAC calculates point thresholding automatic setting method in fundamental matrix |
| CN105184852B (en) * | 2015-08-04 | 2018-01-30 | 百度在线网络技术（北京）有限公司 | A kind of urban road recognition methods and device based on laser point cloud |
| CN107742091B (en) * | 2016-08-22 | 2019-01-29 | 腾讯科技（深圳）有限公司 | A kind of method and device that road shoulder extracts |
| CN106570823B (en) * | 2016-10-11 | 2019-06-18 | 山东科技大学 | Based on matched thick joining method of cloud of plane characteristic |
| CN107610174B (en) * | 2017-09-22 | 2021-02-05 | 深圳大学 | Robust depth information-based plane detection method and system |
-
2018
- 2018-10-30 CN CN201811272869.6A patent/CN109359614B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105551082A (en) * | 2015-12-02 | 2016-05-04 | 百度在线网络技术（北京）有限公司 | Method and device of pavement identification on the basis of laser-point cloud |
| CN106408604A (en) * | 2016-09-22 | 2017-02-15 | 北京数字绿土科技有限公司 | Filtering method and device for point cloud data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109359614A (en) | 2019-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109359614B (en) | Laser point cloud plane identification method, device, equipment and medium | |
| CN106951847B (en) | Obstacle detection method, apparatus, device and storage medium | |
| CN109188457B (en) | Object detection frame generation method, device, equipment, storage medium and vehicle | |
| CN112634181B (en) | Method and device for detecting ground point cloud points | |
| CN109343061B (en) | Sensor calibration method and device, computer equipment, medium and vehicle | |
| CN108629231B (en) | Obstacle detection method, apparatus, device and storage medium | |
| CN109188438B (en) | Yaw angle determination method, device, equipment and medium | |
| CN109190573B (en) | Ground detection method applied to unmanned vehicle, electronic equipment and vehicle | |
| CN112613424A (en) | Rail obstacle detection method, rail obstacle detection device, electronic apparatus, and storage medium | |
| CN112863187B (en) | Detection method of perception model, electronic equipment, road side equipment and cloud control platform | |
| CN112650300B (en) | Unmanned aerial vehicle obstacle avoidance method and device | |
| WO2023040737A1 (en) | Target location determining method and apparatus, electronic device, and storage medium | |
| CN112558035B (en) | Method and device for estimating the ground | |
| CN111724429B (en) | Ground characteristic extraction method and device | |
| CN114549961B (en) | Target object detection method, device, equipment and storage medium | |
| CN112101069A (en) | Method and device for determining driving area information | |
| CN113313654B (en) | Laser point cloud filtering denoising method, system, equipment and storage medium | |
| CN116466704A (en) | Obstacle avoidance optimization method, device, equipment and storage medium | |
| CN115527187A (en) | Method and device for classifying obstacles | |
| CN115937950A (en) | Multi-angle face data acquisition method, device, equipment and storage medium | |
| CN114662600A (en) | Lane line detection method and device and storage medium | |
| CN112906519A (en) | Vehicle type identification method and device | |
| CN113763260B (en) | Water drop noise-based noise filtering method, device, equipment and storage medium | |
| CN115431968B (en) | Vehicle controller, vehicle and vehicle control method | |
| CN110610514A (en) | Method, device and electronic equipment for realizing multi-target tracking |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211021 Address after: 105 / F, building 1, No. 10, Shangdi 10th Street, Haidian District, Beijing 100085 Patentee after: Apollo Intelligent Technology (Beijing) Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TR01 | Transfer of patent right |