CN115728781A - Small obstacle detection method and device based on laser radar point cloud - Google Patents
Small obstacle detection method and device based on laser radar point cloud Download PDFInfo
- Publication number
- CN115728781A CN115728781A CN202110981977.6A CN202110981977A CN115728781A CN 115728781 A CN115728781 A CN 115728781A CN 202110981977 A CN202110981977 A CN 202110981977A CN 115728781 A CN115728781 A CN 115728781A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- segmentation
- ground
- point
- small obstacle
- 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.)
- Pending
Links
Images
Classifications
-
- 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
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Computer Networks & Wireless Communication (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Electromagnetism (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention provides a small obstacle detection method based on laser radar point cloud, which comprises the following steps: extracting gradient features and height features of the point cloud, and performing first segmentation based on the gradient features and the height features to preliminarily distinguish the ground point cloud and the non-ground point cloud; performing a second segmentation on the result of the first segmentation based on at least one of a subdivided region, intensity, and density of the point cloud; performing clustering processing on the point cloud after the second segmentation; and estimating the shape and orientation of the small obstacle based on the result of the clustering process.
Description
Technical Field
The invention relates to the field of laser radars, in particular to a small obstacle detection method based on laser radar point cloud.
Background
In the automatic driving of vehicle scenes such as mine cards, automobiles, locomotives, urban rails and the like, a sensing module of the vehicle serves as the eye of the whole automatic driving system, and reliable environmental information is provided for the automatic driving system. In order to ensure the safety of the vehicle during driving, the sensing module needs to accurately identify small obstacles in front of the vehicle, especially obstacles below 30cm, which are easy to cause damage to tires of the vehicle and even cause tire burst accidents, thus seriously endangering the safety of the vehicle and passengers.
Under the condition of sufficient illumination conditions, the camera can detect the existence of the small obstacle, but the distance precision for detecting the small obstacle through the camera is limited, and the precise position of the small obstacle cannot be accurately obtained. And in the case of insufficient lighting conditions, the camera cannot function effectively. Therefore, in an autonomous driving system, the perception of small obstacles is more dependent on high-resolution lidar.
However, under the influence of complex road conditions such as road surface fluctuation and a ramp, a general radar point cloud segmentation method cannot effectively separate small obstacles from ground point cloud, false reports and false reports are easily generated, so that a laser radar cannot effectively and accurately detect the shapes and the positions of the small obstacles, an effective and accurate environmental information reference cannot be provided for vehicle driving, and the driving safety of automatic driving is threatened.
In order to overcome the above defects in the prior art, there is an urgent need in the art for a method for detecting small obstacles based on laser radar point cloud, which is used for remotely and stably detecting small obstacles in complex scenes such as undulating road surfaces or non-paved road surfaces, for example, in mining areas, rail traffic scenes, and the like, so as to provide effective and accurate environmental information reference for vehicle driving and guarantee driving safety of automatic driving.
Disclosure of Invention
The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.
In order to overcome the defects in the prior art, the invention provides a small obstacle detection method based on laser radar point cloud, which comprises the following steps: extracting gradient features and height features of the point cloud, and performing first segmentation based on the gradient features and the height features to preliminarily distinguish the ground point cloud and the non-ground point cloud; performing a second segmentation on the result of the first segmentation based on at least one of a subdivided region, intensity, and density of the point cloud; performing clustering processing on the point cloud after the second segmentation; and estimating the shape and orientation of the small obstacle based on the result of the clustering process.
In one embodiment, preferably, the second segmentation is performed based on the subdivided regions of the point cloud, including: dividing the point cloud subjected to the first segmentation into a plurality of subdivided regions, wherein the plurality of subdivided regions comprise a near-ground region, a region higher than the ground region and a region lower than the ground region; and setting different gradient thresholds for different subdivided regions to perform the second segmentation.
In an embodiment, preferably, different gradient thresholds are set for different subdivided regions to perform the second segmentation, including: setting a relaxed gradient threshold at above-ground and below-ground regions to perform a second segmentation; and setting a fine gradient threshold in the near-surface region to perform a second segmentation.
In an embodiment, preferably, the second segmentation is performed based on the intensity of the point cloud, including: searching seed points in the point cloud subjected to the first segmentation according to the intensity of the single points and the average intensity of the surrounding neighborhoods of the single points; and judging whether the gradient characteristics of the seed points meet the preset threshold value requirement or not so as to execute the second segmentation.
In one embodiment, preferably, the second segmentation is performed based on the density of the point cloud, including: searching the surrounding neighborhood of the ground point cloud subjected to the first segmentation; and calculating the distance between the points in the surrounding neighborhood of each ground point cloud, and segmenting the point cloud cluster of which the distance between the points is less than a preset threshold value into non-ground point clouds.
In an embodiment, before extracting the gradient feature and the height feature of the point cloud, it is preferable to further include: preprocessing point cloud data; and extracting gradient and height characteristics of the preprocessed point cloud, wherein the preprocessed point cloud data comprises: the point cloud data is matrixed, and the number of rows and columns of the matrixing is determined by the following formula:
wherein, verAngleRange is the range of the vertical angle of the laser radar, verresolution is the resolution of the vertical angle, rows is the number of rows of the matrix, horLangleRange is the range of the horizontal angle of the laser radar, horresolution is the resolution of the horizontal angle, and cols is the number of columns of the matrix.
In an embodiment, preferably, the preprocessing the point cloud data further includes: meanwhile, verresolution and Horresolution are increased in an equal ratio for down sampling so as to improve the operation efficiency.
In one embodiment, preferably, extracting the gradient feature of the point cloud comprises: the gradient of the point cloud P is calculated by the following formula:
wherein alpha is the gradient of the point cloud, and if M and P points are respectively the adjacent lines of the point cloud depth map which belong to the same row as the point cloud, then z is p And z m Respectively corresponding to the M point and the P point and corresponding to the coordinate l in the Z-axis direction p And l m The distances from the coordinate origin O to the projection point of the M point and the P point on the XOY plane are respectively.
In one embodiment, preferably, extracting height features of the point cloud comprises: selecting fitting points according to the value of the point cloud in the Z direction by adopting a random sampling fitting algorithm; judging the fluctuation degree of the ground according to the fitting point, acquiring a ground equation and calculating a segmentation threshold; and acquiring the height characteristic of each point from the ground according to a ground equation and a segmentation threshold value to separate a ground point cloud and a non-ground point cloud.
In an embodiment, preferably, the clustering process is performed on the point cloud after the second segmentation, and includes: and clustering and dividing the non-ground point cloud subjected to the second segmentation by using a density clustering algorithm to obtain a plurality of clustering categories.
In one embodiment, preferably, estimating the shape and orientation of the small obstacle based on the result of the clustering process includes: searching a minimum bounding box for the point cloud cluster of each clustering category, wherein the minimum bounding box wraps all point clouds of the point cloud cluster; and determining the shape and orientation of the small obstacle based on the minimum bounding box.
In one embodiment, preferably, the minimum bounding box is a rectangular bounding box; determining the shape and orientation of the small obstacle based on the minimum bounding box, comprising: the length and width of the rectangular bounding box are the length and width of the small obstacle, and the orientation of the rectangular bounding box is the orientation of the small obstacle.
Another aspect of the present invention provides a small obstacle detection apparatus based on a lidar point cloud, including: a memory; and a processor coupled to the memory, the processor configured to perform the steps of any of the above-described small obstacle detection methods.
The present invention also provides a computer readable medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the above-described small obstacle detection methods.
Drawings
The above features and advantages of the present disclosure will be better understood upon reading the detailed description of embodiments of the disclosure in conjunction with the following drawings. In the drawings, components are not necessarily drawn to scale, and components having similar relative characteristics or features may have the same or similar reference numerals.
Fig. 1 is a schematic flow chart of a method for detecting a small obstacle based on a lidar point cloud according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a computing principle of a point cloud gradient feature according to an embodiment of the invention;
FIG. 3 is a schematic diagram illustrating a method for estimating a minimum bounding box in a clustering process according to an embodiment of the invention; and
fig. 4 is a schematic structural diagram of an apparatus for detecting a small obstacle based on lidar point cloud according to an embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure. While the invention will be described in connection with the preferred embodiments, there is no intent to limit the features of the invention to those embodiments. On the contrary, the invention is described in connection with the embodiments for the purpose of covering alternatives or modifications that may be extended based on the claims of the present invention. In the following description, numerous specific details are included to provide a thorough understanding of the invention. The invention may be practiced without these particulars. Moreover, some of the specific details have been left out of the description in order to avoid obscuring or obscuring the focus of the present invention.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Additionally, the terms "upper," "lower," "left," "right," "top," "bottom," "horizontal," "vertical" and the like as used in the following description are to be understood as referring to the segment and the associated drawings in the illustrated orientation. The relative terms are used for convenience of description only and do not imply that the described apparatus should be constructed or operated in a particular orientation and therefore should not be construed as limiting the invention.
It will be understood that, although the terms first, second, third, etc. may be used herein to describe various elements, regions, layers and/or sections, these elements, regions, layers and/or sections should not be limited by these terms, but rather are used to distinguish one element, region, layer and/or section from another element, region, layer and/or section. Thus, a first component, region, layer or section discussed below could be termed a second component, region, layer or section without departing from some embodiments of the present invention.
In order to overcome the defects in the prior art, the invention provides a small obstacle detection method based on laser radar point cloud, which is used for stably detecting small obstacles under complex road conditions such as ground fluctuation and slopes, so that effective and accurate environmental information reference is provided for vehicle driving, and the driving safety of automatic driving is guaranteed.
Fig. 1 is a schematic method flow diagram of a lidar point cloud-based small obstacle detection method according to an embodiment of the invention.
Referring to fig. 1, a method 100 for detecting a small obstacle based on a laser radar point cloud provided by the present invention includes:
step 101: extracting gradient features and height features of the point cloud, and performing a first segmentation based on the gradient features and the height features to preliminarily distinguish the ground point cloud from the non-ground point cloud.
Due to the characteristic of the point cloud data being disorder in space, in an embodiment, before extracting the gradient feature and the height feature of the point cloud, the method further includes: preprocessing point cloud data; and extracting gradient and height characteristics of the preprocessed point cloud, wherein the preprocessed point cloud data comprises: and performing matrixing on the point cloud data so as to facilitate subsequent segmentation and clustering operations on the point cloud.
The number of rows and columns of the matrixing is determined by the following formula:
wherein, verAngleRange is the range of the vertical angle of the laser radar, verResolution is the resolution of the vertical angle, rows is the number of rows of the matrix, horAngleRange is the range of the horizontal angle of the laser radar, horResolution is the resolution of the horizontal angle, and cols is the number of columns of the matrix.
In one embodiment, in order to improve the operation efficiency of the algorithm and reduce the number of point searches, down-sampling is performed by increasing two values, verResolution and HorResolution. And when the Verresolution value and the Horresolution value are increased, geometric increase is adopted to ensure that the shape of the outline of the obstacle is not deformed.
After the point cloud data is preprocessed, executing step 101: extracting gradient features and height features of the point cloud, and performing a first segmentation based on the gradient features and the height features to preliminarily distinguish the ground point cloud from the non-ground point cloud.
Fig. 2 is a schematic diagram illustrating a computing principle of a point cloud gradient feature according to an embodiment of the invention.
After the point cloud is matrixed, the features between the beams are the features between the matrix rows, the features in the beams are the features between the columns, and the key to the segmentation of the ground point cloud and the non-ground point cloud is the change of the longitudinal gradient, as shown in fig. 2, the change of the gradient can be represented by an angle α.
Referring to FIG. 2, M and P points are respectively corresponding to two point clouds z belonging to adjacent rows in the same row in the point cloud depth map p And z m Respectively corresponding to the M point and the P point and corresponding to the coordinate l in the Z-axis direction p And l m The distances from the coordinate origin O to the projection point of the M point and the P point on the XOY plane are respectively. l p And l m Calculated by the following formula:
wherein x is p And y p X-axis coordinate, Y-axis coordinate, X of P point m And y m Respectively an X-axis coordinate and a Y-axis coordinate of the M point.
The gradient of the point cloud is calculated by the following formula:
wherein α is the gradient of the point cloud.
Because the angle alpha of the ground point cloud and the obstacle point cloud is greatly different, the ground point cloud and the non-ground point cloud can be distinguished according to the gradient feature alpha of the point cloud. The threshold value alpha can be selected according to the actual debugging experience value 0 When alpha < alpha is satisfied 0 Judging the point cloud positioned in the current row and column in the depth map as a ground point; otherwise, alpha < alpha is not satisfied 0 And judging the non-ground point as the non-ground point.
Besides the gradient feature of the point cloud, the height feature of the point cloud also needs to be extracted. The height feature characterizes the height of each point from the ground, also referred to as a ground feature.
In an embodiment, the ground feature extraction process is implemented by using a block plane fitting method, and a Random Sample Consensus (RANSAC) algorithm is adopted in the plane fitting process.
The algorithm is to estimate a mathematical model from a set of observed data in an iterative manner. The RANSAC algorithm assumes that the data contains both correct data and anomalous data. The correct data is called the inner point and the outlier is called the outer point. At the same time the RANSAC algorithm also assumes that, given a correct set of data, there are methods by which the model parameters that fit these data can be calculated.
After the original point cloud is matrixed, fitting points of RANSAC are roughly selected according to values of the point cloud in the z direction in a segmented and regionalized mode, the fluctuation degree of the ground is judged according to the total number of the fitting points occupied by the inner points and the outer points, and a ground equation and a self-adaptive calculation ground segmentation threshold value are obtained. According to the regional ground equation and the ground segmentation threshold, non-ground points of each region can be separated, and the height characteristics of each point from the ground can be obtained according to the point-to-plane equation, so that ground point cloud and non-ground point cloud can be separated.
Referring back to fig. 1, the method 100 for detecting a small obstacle based on lidar point cloud further includes:
step 102: performing a second segmentation on the result of the first segmentation based on at least one of the subdivided region, intensity, and density of the point cloud.
The second segmentation is to re-segment the point clouds segmented by mistake in the first segmentation on the basis of the first segmentation, so that the number of the point clouds reserved by the small obstacles is larger, and the small obstacles can be detected at a longer distance.
The laser radar point cloud-based small obstacle detection method provided by the invention comprises at least one of the following two segmentation steps aiming at ground point cloud and non-ground point cloud: firstly, based on the characteristics obtained by the first segmentation, based on the subdivision region of the point cloud, different thresholds are adopted in the fixed region for re-segmentation; secondly, extracting the obstacle point cloud based on the intensity of the single point cloud and the gradient of the surrounding neighborhood; thirdly, extracting the point cloud of the small obstacle based on the density characteristics of the point cloud.
And performing second segmentation on the point cloud-based subdivided regions, namely dividing the point cloud subjected to the first segmentation into a plurality of subdivided regions, wherein the plurality of subdivided regions comprise a near-ground region, a region higher than the ground region and a region lower than the ground region. Through different gradient threshold values of point clouds in different areas, the point clouds of the near-ground target are extracted more finely, and the small target can be divided to obtain more point clouds. Point clouds above or below the ground are generally large obstacles and can be segmented using more relaxed thresholds, with fine gradient thresholds being set in the near-ground region to perform a second segmentation.
In one embodiment, the threshold of the gradient of the point cloud above or below the ground is 0.8, and the threshold of the gradient of the target point cloud near the ground is 0.12, it is understood that the threshold of the gradient of the point cloud above or below the ground is more relaxed, and a larger value can be set, for example, in the range of 0.5 to 1. The gradient of the near-surface point cloud is required to be finer, and a smaller threshold value needs to be selected, for example, the range of 0.1 to 0.5, so that obstacles for dividing different sub-divided areas are finer and targeted.
The point cloud intensity refers to the light intensity of a single point reflection received by the laser radar. Generally, the point cloud intensity of the obstacle will be significantly higher than the intensity of the ground point cloud. The intensity information of the laser radar is not accurate, so that point clouds with prominent single-point intensity are considered and searched in the surrounding neighborhood.
In an embodiment, the surrounding neighborhood of the point cloud is selected to be 25 neighborhoods of 5 × 5, and a neighborhood range can be flexibly selected according to different resolutions of the laser radar in an actual working condition.
In an embodiment, performing the second segmentation based on the intensity of the point cloud comprises: and searching seed points in the ground point cloud after the first segmentation and the area segmentation according to the average intensity of the neighborhood around the point cloud and the intensity of the single point. After the seed point is found, the gradient characteristics of the points of the surrounding neighborhood are searched, whether the gradient characteristics meet the threshold requirement or not is judged, if yes, the point cloud of the region is segmented into non-ground point cloud, and therefore a more accurate segmentation result is obtained.
According to the scanning mode of the laser radar, the neighborhood point clouds of the non-ground point clouds are denser. In one embodiment, performing the second segmentation based on the density of the point cloud comprises: and searching surrounding neighborhoods of the ground point clouds subjected to the first segmentation, calculating the distance between a point and a neighboring neighborhood of each ground point cloud, and segmenting the point cloud cluster into non-ground point clouds if a certain point neighborhood contains a plurality of point clouds with closer distances.
On the basis of the first point cloud segmentation, the invention provides the second segmentation based on at least one of the subdivision region, the strength and the density of the point cloud, so that small obstacles can be more accurately and effectively identified, and the environmental requirements under complex road condition scenes such as ground fluctuation, ramps and the like are met.
Referring back to fig. 1, the method 100 for detecting a small obstacle based on a lidar point cloud further includes:
step 103: and performing clustering processing on the point cloud after the second segmentation.
And performing clustering processing on the point cloud after the second segmentation, namely performing clustering division on the non-ground point cloud after the second segmentation by using a density clustering algorithm to obtain a plurality of clustering categories.
After the ground segmentation and the region of interest extraction, non-ground point cloud in the region of interest is obtained. In order to obtain obstacle information such as a vehicle from the point cloud, it is necessary to further perform clustering processing on the point cloud of the region of interest. The primary purpose of the clustering algorithm is to divide the dispersed lidar point cloud into several independent point cloud sets. And taking the obtained point cloud set as an obstacle.
In an embodiment, the clustering process employs the DBSCAN algorithm. DBSCAN is a density-based clustering algorithm that is generally assumed to be determined by how closely the samples are distributed. Samples of the same class are closely connected to each other, so that samples of the same class must exist in a short distance around any sample of the class. By classifying closely connected samples into one class, a cluster type is obtained, and the number of obstacles of non-ground points and corresponding point cloud clusters are obtained.
Referring back to fig. 1, the method 100 for detecting a small obstacle based on a lidar point cloud further includes:
step 104: the shape and orientation of the small obstacle are estimated based on the result of the clustering process.
In order to obtain the state of the obstacle, such as size, shape, etc., shape estimation needs to be performed on the clustered point cloud.
In one embodiment, the shape and orientation of the small obstacle are estimated based on the result of the clustering process, a minimum bounding box is found for the point cloud cluster of each clustering class, and the minimum bounding box wraps all the point clouds of the point cloud cluster; and determining the shape and the orientation of the small obstacle based on the minimum bounding box.
Fig. 3 is a schematic diagram illustrating a method for estimating a minimum bounding box in a clustering process according to an embodiment of the invention.
Referring to fig. 3, the minimum border estimation method searches for a minimum rectangular border frame capable of wrapping all the point clouds of each group under a certain angle for each group of clustered laser point clouds, the possible direction θ of the rectangle is in the range of 0 to 90 °, only a single side between 0 to 90 ° is considered because the adjacent two sides of the rectangle are orthogonal, the direction of the other side is θ + pi/2, and the degree of fitting of the rectangular frame is defined by the distance between the clustering point and the two sides of the right angle:
wherein i is defined as the ith point in the m points of the cluster of cluster points, d i Defined as the distance from the ith point to the nearest edge of the rectangle fitting box.
And calculating the sum of reciprocal distances from all point clouds to the boundary in the boundary frame, changing the angle, and repeating the steps until all angles from 0 to 90 degrees are traversed. The reciprocal and the maximum bounding box are the required minimum bounding box, the length and the width of the rectangular bounding box are the length and the width of the small obstacle, and the orientation of the rectangular bounding box is the orientation of the small obstacle.
While, for purposes of simplicity of explanation, the methodologies are shown and described as a series of acts, it is to be understood and appreciated that the methodologies are not limited by the order of acts, as some acts may, in accordance with one or more embodiments, occur in different orders and/or concurrently with other acts from that shown and described herein or not shown and described herein, as would be understood by one skilled in the art.
Fig. 4 is a schematic structural diagram of a device for detecting a small obstacle based on a lidar point cloud according to an embodiment of the invention.
As shown in fig. 4, the computer system/server 400 of the small obstacle detecting apparatus is represented in the form of a general-purpose computer device. The components of the computer system/server 400 may include one or more processors 402, memory 401, and a bus 403 that connects the various system components (including the memory 401 and the processors 402).
The bus 403 includes a data bus, an address bus, and a control bus. The product of the number of bits of the data bus and the operating frequency is proportional to the data transfer rate, the number of bits of the address bus determines the maximum addressable memory space, and the control bus (read/write) indicates the type of bus cycle and the time at which the present I/O operation is completed. The processor 402 is connected to the memory 401 via a bus 403 and is configured to implement the vehicle control method provided by any of the above embodiments.
The processor 402 is a final execution unit for information processing and program execution, which is an operation and control core of the computer system/server 400 as the small obstacle detection device. The operation of all software layers in the computer system will ultimately be mapped to the operation of the processor 402 by the instruction set. The processor 402 has the main functions of processing instructions, executing operations, controlling time and processing data.
The memory 401 is a variety of storage devices for storing programs and data in the computer. Memory 401 may include computer system readable media in the form of storage volatile memory. Such as Random Access Memory (RAM) 404 and/or cache memory 405.
A Random Access Memory (RAM) 404 is an internal memory that exchanges data directly with the processor 402. It can be read and written at any time (except for refreshing), and is fast, usually used as a temporary data storage medium for an operating system or other programs in operation, and the stored data will be lost when power is off. Cache memory (Cache) 405 is a level one memory existing between main memory and processor 402, and has a relatively small capacity but much higher speed than main memory, close to the speed of processor 402.
The computer system/server 400 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. In this embodiment, the storage system 406 may be used to read from and write to non-removable, nonvolatile magnetic media.
Computer system/server 400 may also communicate with one or more external devices 408 (e.g., keyboard, pointing device, display 409, etc.), with one or more devices that enable a user to interact with the computer system/server 400, and/or with any devices (e.g., network card, modem, etc.) that enable the computer system/server 400 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interfaces 410.
Computer system/server 400 may also 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) via network adapter 411. As shown in FIG. 4, network adapter 411 communicates with the other modules of computer system/server 400 via bus 403. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with computer system/server 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, to name a few.
The present invention also provides an embodiment of a computer-readable medium having a computer program stored thereon. When being executed by a processor, the computer program can realize the steps of any one of the small obstacle detection methods based on the laser radar point cloud.
Those of skill in the art would understand that information, signals, and data may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits (bits), symbols, and chips that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.
Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The processors described herein may be implemented using electronic hardware, computer software, or any combination thereof. Whether such processors are implemented as hardware or software depends upon the particular application and the overall design constraints imposed on the system. As an example, a processor, any portion of a processor, or any combination of processors presented in this disclosure may be implemented with a microprocessor, a microcontroller, a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a state machine, gated logic, discrete hardware circuitry, and other suitable processing components configured to perform the various functions described throughout this disclosure. The functionality of a processor, any portion of a processor, or any combination of processors presented in this disclosure may be implemented in software executed by a microprocessor, microcontroller, DSP, or other suitable platform.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.
In one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a web site, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk (disk) and disc (disc), as used herein, includes Compact Disc (CD), laser disc, optical disc, digital Versatile Disc (DVD), floppy disk and blu-ray disc where disks (disks) usually reproduce data magnetically, while discs (discs) reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (14)
1. A small obstacle detection method based on laser radar point cloud comprises the following steps:
extracting gradient features and height features of the point cloud, and performing a first segmentation based on the gradient features and the height features to preliminarily distinguish a ground point cloud from a non-ground point cloud;
performing a second segmentation on the result of the first segmentation based on at least one of subdivided regions, intensity, and density of the point cloud;
performing clustering processing on the point cloud after the second segmentation; and
estimating a shape and orientation of the small obstacle based on a result of the clustering process.
2. The small obstacle detection method according to claim 1, wherein performing a second segmentation based on the subdivided regions of the point cloud comprises:
dividing the point cloud subjected to the first segmentation into a plurality of subdivided regions, wherein the plurality of subdivided regions comprise a near-ground region, a region higher than the ground region and a region lower than the ground region; and
different gradient thresholds are set for different subdivided regions to perform the second segmentation.
3. The small obstacle detection method according to claim 2, wherein the setting of different gradient thresholds for different subdivided regions to perform the second segmentation comprises:
setting relaxed said gradient threshold at said above ground region and said below ground region to perform said second segmentation; and
setting the fine gradient threshold in the near-surface region to perform the second segmentation.
4. The small obstacle detection method of claim 1, wherein performing a second segmentation based on the intensity of the point cloud comprises:
searching a seed point in the point cloud subjected to the first segmentation according to the intensity of the single point and the average intensity of the surrounding neighborhood of the single point; and
and judging whether the gradient characteristics of the seed points meet the requirement of a preset threshold value or not so as to execute the second segmentation.
5. The small obstacle detection method of claim 1, wherein performing a second segmentation based on the density of the point cloud comprises:
searching the surrounding neighborhood of the ground point cloud after the first segmentation; and
and calculating the distance between the points in the surrounding neighborhood of each ground point cloud, and segmenting the point cloud cluster of which the distance between the points is less than a preset threshold value into non-ground point clouds.
6. The small obstacle detection method of claim 1, further comprising, prior to the extracting gradient features and height features of the point cloud:
preprocessing point cloud data; and
extracting gradient and height characteristics of the preprocessed point cloud, wherein the preprocessed point cloud data comprises:
matrixing the point cloud data, wherein the number of rows and columns of the matrixing is determined by the following formula:
wherein, verAngleRange is the range of the vertical angle of the laser radar, verresolution is the resolution of the vertical angle, rows is the number of rows of the matrix, horLangleRange is the range of the horizontal angle of the laser radar, horresolution is the resolution of the horizontal angle, and cols is the number of columns of the matrix.
7. The small obstacle detection method of claim 6, wherein the preprocessing point cloud data, further comprises:
meanwhile, verresolution and Horresolution are increased in an equal ratio for down sampling so as to improve the operation efficiency.
8. The small obstacle detection method according to claim 1, wherein the extracting gradient features of the point cloud comprises:
the gradient of the point cloud is calculated by the following formula:
wherein alpha is the gradient of the point cloud, and if M and P points are respectively the adjacent lines of the point cloud depth map which belong to the same column with the point cloud, then z is p And z m Respectively corresponding to the M point and the P point and corresponding to the coordinate l in the Z-axis direction p And l m Respectively, M point and P point areThe distance of the projected point of the XOY plane from the origin of coordinates O.
9. The small obstacle detection method of claim 1, wherein extracting the height features of the point cloud comprises:
selecting fitting points according to the value of the point cloud in the Z direction by adopting a random sampling fitting algorithm;
judging the fluctuation degree of the ground according to the fitting point, acquiring a ground equation and calculating a segmentation threshold; and
and acquiring the height characteristic of each point from the ground according to the ground equation and a segmentation threshold value to separate the ground point cloud and the non-ground point cloud.
10. The small obstacle detection method according to claim 1, wherein the performing a clustering process on the point cloud after the second segmentation includes:
and clustering and dividing the non-ground point cloud subjected to the second segmentation by using a density clustering algorithm to obtain a plurality of clustering categories.
11. The small obstacle detection method according to claim 10, wherein the estimating the shape and orientation of the small obstacle based on the result of the clustering process includes:
searching a minimum bounding box for the point cloud cluster of each clustering category, wherein the minimum bounding box wraps all point clouds of the point cloud cluster; and
determining a shape and orientation of the small obstacle based on the minimum bounding box.
12. The small obstacle detection method according to claim 11, wherein the minimum bounding box is a rectangular bounding box;
the determining the shape and orientation of the small obstacle based on the minimum bounding box comprises:
the length and the width of the rectangular bounding box are the length and the width of the small obstacle, and the orientation of the rectangular bounding box is the orientation of the small obstacle.
13. A small obstacle detection device based on laser radar point cloud includes:
a memory; and
a processor coupled to the memory, the processor configured to perform the steps of the small obstacle detection method of any one of claims 1 to 12.
14. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method for detecting small obstacles according to any one of claims 1 to 12.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110981977.6A CN115728781A (en) | 2021-08-25 | 2021-08-25 | Small obstacle detection method and device based on laser radar point cloud |
| PCT/CN2021/126726 WO2023024241A1 (en) | 2021-08-25 | 2021-10-27 | Laser radar point cloud-based method and apparatus for detecting small obstacle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110981977.6A CN115728781A (en) | 2021-08-25 | 2021-08-25 | Small obstacle detection method and device based on laser radar point cloud |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115728781A true CN115728781A (en) | 2023-03-03 |
Family
ID=85289665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110981977.6A Pending CN115728781A (en) | 2021-08-25 | 2021-08-25 | Small obstacle detection method and device based on laser radar point cloud |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115728781A (en) |
| WO (1) | WO2023024241A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115973131B (en) * | 2023-03-20 | 2023-06-13 | 上海伯镭智能科技有限公司 | Mining area unmanned vehicle rollover prevention method and related device |
| CN117670822A (en) * | 2023-12-05 | 2024-03-08 | 北京路凯智行科技有限公司 | Method and system for detecting bumpiness of non-hardened road surface, electronic device and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101762504B1 (en) * | 2015-08-31 | 2017-07-28 | 고려대학교 산학협력단 | Method for detecting floor obstacle using laser range finder |
| CN109633686B (en) * | 2018-11-22 | 2021-01-19 | 浙江中车电车有限公司 | Method and system for detecting ground obstacle based on laser radar |
| CN110674705B (en) * | 2019-09-05 | 2022-11-29 | 北京智行者科技股份有限公司 | Small-sized obstacle detection method and device based on multi-line laser radar |
| CN111192284B (en) * | 2019-12-27 | 2022-04-05 | 吉林大学 | Vehicle-mounted laser point cloud segmentation method and system |
| CN112529874A (en) * | 2020-12-14 | 2021-03-19 | 上海智蕙林医疗科技有限公司 | Obstacle detection method, device, medium and robot based on three-dimensional radar |
-
2021
- 2021-08-25 CN CN202110981977.6A patent/CN115728781A/en active Pending
- 2021-10-27 WO PCT/CN2021/126726 patent/WO2023024241A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023024241A1 (en) | 2023-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200410690A1 (en) | Method and apparatus for segmenting point cloud data, storage medium, and electronic device | |
| CN106951879B (en) | Multi-feature fusion vehicle detection method based on camera and millimeter wave radar | |
| CN113874927B (en) | Parking detection method, system, processing device and storage medium | |
| WO2018068653A1 (en) | Point cloud data processing method and apparatus, and storage medium | |
| CN106570446A (en) | Lane line extraction method and device | |
| CN107169464A (en) | A kind of Method for Road Boundary Detection based on laser point cloud | |
| Daigavane et al. | Road lane detection with improved canny edges using ant colony optimization | |
| CN109583280A (en) | Lane detection method, apparatus, equipment and storage medium | |
| CN107192994A (en) | Multi-line laser radar mass cloud data is quickly effectively extracted and vehicle, lane line characteristic recognition method | |
| CN108845569A (en) | Generate semi-automatic cloud method of the horizontal bend lane of three-dimensional high-definition mileage chart | |
| CN104573627A (en) | Lane line reservation and detection algorithm based on binary image | |
| CN111488808B (en) | Lane line detection method based on traffic violation image data | |
| CN115728781A (en) | Small obstacle detection method and device based on laser radar point cloud | |
| CN109584294A (en) | A kind of road surface data reduction method and apparatus based on laser point cloud | |
| CN107705577B (en) | Real-time detection method and system for calibrating illegal lane change of vehicle based on lane line | |
| EP4120123A1 (en) | Scan line-based road point cloud extraction method | |
| CN114488194A (en) | Method for detecting and identifying targets under structured road of intelligent driving vehicle | |
| CN114842450A (en) | Driving region detection method, device and equipment | |
| Hu | Intelligent road sign inventory (IRSI) with image recognition and attribute computation from video log | |
| Wei et al. | Damage inspection for road markings based on images with hierarchical semantic segmentation strategy and dynamic homography estimation | |
| WO2020170745A1 (en) | Traveling road recognition device | |
| JPWO2020090388A1 (en) | Map generation system, map generation method and map generation program | |
| KR101347886B1 (en) | Method and Apparatus for Road Lane Recognition by Surface Region and Geometry Information | |
| CN111881752A (en) | Guardrail detection and classification method and device, electronic equipment and storage medium | |
| Phan et al. | Pavement crack extraction method from mobile laser scanning point cloud |
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 |