CN111208533A - Real-time ground detection method based on laser radar - Google Patents
Real-time ground detection method based on laser radar Download PDFInfo
- Publication number
- CN111208533A CN111208533A CN202010023358.1A CN202010023358A CN111208533A CN 111208533 A CN111208533 A CN 111208533A CN 202010023358 A CN202010023358 A CN 202010023358A CN 111208533 A CN111208533 A CN 111208533A
- Authority
- CN
- China
- Prior art keywords
- point
- plane
- value
- points
- point cloud
- 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
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/06—Systems determining position data of a target
Abstract
The invention relates to a real-time ground detection method based on a laser radar, which comprises the following steps: s1, removing invalid points in the laser radar data point cloud set; s2, equally dividing a circle in the horizontal direction with a laser radar as the center into a plurality of sector areas, finding out a point with the minimum Z value projected in each sector area, screening out a point with the height difference between the point with the minimum Z value and the corresponding point in each sector area in a set range, and classifying the point into a point cloud set to be fitted; s3, performing plane fitting on the point cloud set to be fitted to obtain a plane equation; and S4, performing cycle traversal on the point cloud set obtained in the step S1, and comparing the Z value of the data point with the height of the fitting plane to realize ground detection. Compared with the prior art, the ground detection method can accurately detect the ground in real time in an urban traffic scene.
Description
Technical Field
The invention relates to a laser radar detection method in the field of automatic driving, in particular to a real-time ground detection method based on a laser radar.
Background
With the rapid development of different types of perception sensors such as three-dimensional laser radars, millimeter wave radars and cameras, the perception function of the environment of the automatic driving automobile is gradually mature, so that the automatic driving slowly enters the sight of the public. In an outdoor environment perception scene, ground segmentation of the laser radar is an important preprocessing task of environment perception, is the basis of obstacle detection, classification and dynamic target tracking, and can help to reduce the size of data to be processed and further reduce the overall calculation time. Due to the fact that the three-dimensional laser radar data are distributed unevenly, namely point clouds close to the laser radar are distributed densely relatively, point clouds far away from the laser radar are distributed sparsely relatively, and therefore the phenomena of over-segmentation, under-segmentation, slow segmentation and the like often occur during ground segmentation in a complex urban traffic environment.
The main task of ground segmentation is to distinguish ground points from obstacles on the ground, because when vehicles are detected on both the left and right sides of the lidar during driving, the presence of the ground points may connect the two vehicles together, which will affect the detection of the obstacles. When the ground points are removed, the point cloud becomes a single individual in space without any connection, which is beneficial to the operation of target detection such as grid classification, obstacle detection and the like. The existing ground detection method comprises the following steps: the method is based on three-dimensional point cloud projection, a plane is fitted based on point cloud neighborhood information, a region growing algorithm is adopted, and a point cloud model is segmented based on methods such as surface element classification, mean shift clustering and spectral clustering.
The three-dimensional point cloud is projected to a horizontal grid, and the height threshold of the point cloud in the adjacent grid is determined by comparing and determining the difference of the attributes of the laser radar points, but the method only calculates the height difference between the laser point clouds, has high efficiency, and easily causes the problem of insufficient segmentation.
And a local plane is fitted according to neighborhood information of the point cloud, and the point cloud is segmented by using the local convexity of the plane normal vector, so that the segmentation effect on different targets is better, but the convexity characteristic of the rough surface normal vector is not well calculated.
And the point cloud segmentation is carried out by adopting a region growing algorithm, so that the realization is simple and the speed is high. But different region growing strategies often lead to different levels of detail decomposition results.
The point cloud model is segmented based on methods such as surface element classification, mean shift clustering and spectral clustering, and the point cloud is divided into a plurality of groups through feature similarity detection. Improper clustering algorithm design may result in over-segmentation or under-segmentation, since different clustering criteria may produce different clustering results.
Disclosure of Invention
The invention aims to provide a real-time ground detection method based on a laser radar, which aims to solve the problems of over-segmentation, under-segmentation, slow segmentation and the like existing in the existing ground segmentation based on the point cloud of the laser radar.
The purpose of the invention can be realized by the following technical scheme:
a real-time ground detection method based on laser radar comprises the following specific steps:
s1: and selecting the characteristic points of the acquired laser radar point cloud data, and removing invalid points before selecting the characteristic points, thereby being beneficial to characteristic point extraction and plane fitting.
S2: dividing a circle with a laser radar as a center into 180 sector areas by taking 2 degrees as one part, finding out points with the minimum Z value projected in each sector area, screening out points in each sector area, which conform to the range of the minimum Z value point height difference of the sector area, and classifying the points into a point cloud set of plane fitting.
S3: RANSAC plane fitting is carried out on the characteristic points to obtain a plane equation.
S4: and circularly traversing all the points, and comparing the Z values of the points with the height of the fitting plane to realize the detection of the ground.
Two problems need to be considered in the point cloud segmentation algorithm: 1) processing a large amount of point cloud data; 2) how to select feature points. The point clouds meeting the requirements of people are screened out from a large amount of point cloud data, and the operation efficiency of the algorithm can be improved by filtering the interference point clouds. The characteristic points are selected to ensure that the points are uniformly distributed in the scanning range of the laser radar, have high sensitivity and are less influenced by noise, and the selected characteristic points are used for establishing an initial plane model of the ground, so that how to select the initial characteristic points is important.
In S1, for the point cloud data acquired by the laser radar, the existence of the noise point cloud necessarily affects the plane fitting, and therefore needs to be eliminated. The point cloud set P obtained by cycle traversal is { P ═ P1,p2,…,pnObtaining coordinates and intensity values of the point clouds in the point cloud set, and if any one of the coordinate values x, y, z or the intensity values is none or the values of x, y and z are less than 1.e-6, judging that the point is accepted as an invalid point set V ═ V ═ 61,v2,…,viAnd finally obtaining a point cloud set P after the invalid points are removedNV。
The screening of the feature points in the step S2 is mainly based on the height information of the point cloud. In the laser radar point cloud, the ground laser radar point cloud is annular, the distance from the ground laser radar point cloud to the laser radar origin is gradually increased, and the points of the laser radar in the area of the obstacle are in linear distribution. Extracting characteristic points of the point cloud data after the invalid points are removed, dividing a circle with the laser radar as the center into 180 fan-shaped areas by taking 2 degrees, and finding out a point LP _ i with the minimum z value projected in each fan-shaped area, wherein i is {1,2, …,180}, so that the data are not influenced by measurement noise in a plane fitting stage. The screened characteristic point with the minimum z value of each fan-shaped area is used as the lowest height point of each fan-shaped area, other points in each area are compared with the characteristic point, and the characteristic point cloud set FP which accords with plane fitting is classified into the height difference coincidence range between the points1,fp2,…,fpN}。
S3 is mainly performed By performing RANSAC plane fitting on the feature points selected in S2 to obtain a plane parameter and a plane fitting equation Ax + By + Cz + D of 0, and includes the following steps:
s3.1, randomly selecting three characteristic points from the characteristic point cloud set FP obtained in the step S2, ensuring that the three points are not on the same straight line, and obtaining plane parameters from the three characteristic points to obtain a plane equation.
And S3.2, calculating distance parameters from all other characteristic points to the plane, and calculating the number of points in the plane.
S3.3 Loop Steps S3.1, S3.2And obtaining the number of the fitting plane coincidence points each time.
S3.4, reserving an optimal number of plane equations.
And S4 circularly traverses all points except the characteristic points, effective points except the characteristic points are substituted into a plane fitting equation, when the z value of the point and the calculated result are in an error range, the point is classified into the point cloud set of the ground points, otherwise, the point is classified into the non-ground points, and finally, the ground detection is realized.
Compared with the prior art, the invention has the following advantages:
(1) the number of point clouds scanned by a frame of the 16-line laser radar can reach 320000 points at most, so that invalid points are removed before feature points of the acquired laser radar point cloud data are selected, and feature point extraction and plane fitting are facilitated.
(2) The characteristic points are selected to ensure that the points are uniformly distributed in the scanning range of the laser radar and have higher sensitivity and less influence of noise, the circle with the laser radar as the center is divided into 180 fan-shaped areas by taking 2 degrees as one part, and the point with the minimum z value projected in each fan-shaped area is found out, so that the data is not influenced by measurement noise in the plane fitting stage.
Drawings
Fig. 1 is a flowchart of the algorithm of the present embodiment.
Fig. 2 is an original point cloud image of a lidar acquisition.
Fig. 3 is a diagram of real-time surface detection when no moving object is present around the present embodiment.
Fig. 4 is a diagram of real-time surface detection when the present embodiment turns at a traffic intersection.
Fig. 5 is a ground detection view of the present embodiment as the lidar sweeps around many vehicles.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Examples
As shown in figures 1 to 5:
fig. 1 is a method for real-time ground detection based on a laser radar according to this embodiment, which includes the following steps:
1) and scanning urban road environment information around the automatic driving vehicle as a center, and inputting point cloud data scanned by the laser radar in real time.
2) The existence of the invalid point is easy to cause error influence on the extraction of the characteristic point and the fitting of the plane, so that whether the point cloud belongs to the invalid point or not is judged according to the coordinate and the intensity value information of the point cloud and the point cloud is eliminated.
3) And for the point cloud data after the invalid points are removed, dividing the point cloud data into 180 fan-shaped areas by taking the laser radar as a center and taking 2 degrees as one part, and projecting each point into the corresponding fan-shaped area.
4) Screening the projected point clouds to screen out the point P (P) with the minimum Z value in each fan-shaped area1,p2,…,p180}。
5) The remaining points of each sector are compared with the point with the minimum Z value in the area, and the height difference is classified into the point cloud set of the fitting plane according to the range.
6) And randomly selecting three characteristic points from the point cloud set of the characteristic points and ensuring that the three points are not on the same straight line, and obtaining plane parameters and a plane equation by using the three characteristic points. And calculating the distance parameters from all other characteristic points to the plane, and calculating the number of points in the plane. The loop steps S3.1 and S3.2 are sharedAnd obtaining the number of the fitting plane coincidence points each time. And reserving the optimal number of plane equations.
7) And circularly traversing all the points except the characteristic points, bringing the effective points except the characteristic points into a plane fitting equation, and when the z value of the point and the calculated result are in an error range, putting the z value of the point into the point cloud set of the ground points, and otherwise, putting the z value of the point into the non-ground points, thereby finally realizing the detection of the ground.
TABLE 1 pseudo-code based on lidar real-time ground detection
Fig. 2 is a raw point cloud image acquired by a laser radar without any processing of point cloud data.
The laser radar installing support of this embodiment is customized according to the model of car, installs laser radar on the support, ensures the covering surface of its stability and data.
Fig. 3 is a diagram of real-time surface detection when no moving object is present around the present embodiment. The classification of the obstacles and the ground can be clearly seen, white lines represent detected ground data, and other color data represent obstacles such as roadside trees, buildings and the like.
Fig. 4 is a diagram of real-time surface detection when the present embodiment turns at a traffic intersection. The results demonstrate the performance of the algorithm for ground detection when turning.
Fig. 5 is a ground detection view of the present embodiment as the lidar sweeps around many vehicles. The left side of the vehicle body is provided with 6 vehicles, the right side of the vehicle body is provided with 5 vehicles, and the experimental result shows that the ground and obstacle detection effect is good when too many vehicles exist around the laser radar, and the situation that the ground detection result is interfered by obstacles does not exist.
Claims (8)
1. A real-time ground detection method based on laser radar is characterized by comprising the following steps:
s1, removing invalid points in the laser radar data point cloud set;
s2, equally dividing a circle in the horizontal direction with a laser radar as the center into a plurality of sector areas, finding out a point with the minimum Z value projected in each sector area, screening out a point with the minimum Z value corresponding to the point in each sector area, wherein the height difference of the point is in a set range, and classifying the point into a point cloud set to be fitted, wherein the Z value is a height component;
s3, performing plane fitting on the point cloud set to be fitted to obtain a plane equation;
and S4, performing cycle traversal on the point cloud set obtained in the step S1, and comparing the Z value of the data point with the height of the fitting plane to realize ground detection.
2. The method for real-time ground detection based on lidar according to claim 1, wherein in step S1, the invalid point removing method comprises: and circularly traversing the point cloud set to obtain the three-dimensional coordinate value and the intensity value of each data point, and if any one of the coordinate value components and the intensity value is null data or when any one of the coordinate value components is smaller than a set value, judging that the point is an invalid point.
3. The lidar-based real-time ground detection method according to claim 2, wherein the set value is 1 x 10-6。
4. The method according to claim 1, wherein in step S2, the circular arc angle of each sector is 2 °.
5. The lidar-based real-time ground detection method of claim 1, wherein in step S3, RANSAC plane fitting is performed on the cloud set of points to be fitted.
6. The lidar based real-time ground detection method according to claim 5, wherein the step S3 comprises the steps of:
s3.1, randomly selecting three characteristic points which are not on the same straight line from the point cloud set to be fitted obtained in the step S2, and determining a plane according to the three characteristic points;
s3.2, calculating the distances from all other points to the plane to obtain the number of the points in the plane;
s3.3, repeating the steps S3.1 and S3.2 until all planes are traversed to obtain the number of points in each fitting plane;
and S3.4, reserving an optimal plane equation.
7. The method according to claim 6, wherein the optimal plane equation is a plane equation with the largest number of points in a plane.
8. The method according to claim 1, wherein the step S4 specifically comprises: and (5) performing cyclic traversal on the point cloud set obtained in the step (S1), comparing the Z value of the data point with the height of the fitting plane to obtain a difference value, and taking the point of the difference value within the error range as a ground point to realize ground detection.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010023358.1A CN111208533A (en) | 2020-01-09 | 2020-01-09 | Real-time ground detection method based on laser radar |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010023358.1A CN111208533A (en) | 2020-01-09 | 2020-01-09 | Real-time ground detection method based on laser radar |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111208533A true CN111208533A (en) | 2020-05-29 |
Family
ID=70788787
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010023358.1A Pending CN111208533A (en) | 2020-01-09 | 2020-01-09 | Real-time ground detection method based on laser radar |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111208533A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112036274A (en) * | 2020-08-19 | 2020-12-04 | 江苏智能网联汽车创新中心有限公司 | Driving region detection method and device, electronic equipment and storage medium |
CN112180343A (en) * | 2020-09-30 | 2021-01-05 | 东软睿驰汽车技术(沈阳)有限公司 | Laser point cloud data processing method, device and equipment and unmanned system |
CN113050106A (en) * | 2021-02-05 | 2021-06-29 | 上海擎朗智能科技有限公司 | Ground detection method, device, electronic equipment, system and medium |
CN113689329A (en) * | 2021-07-02 | 2021-11-23 | 上海工程技术大学 | Shortest path interpolation method for enhancing sparse point cloud |
CN113744323A (en) * | 2021-08-11 | 2021-12-03 | 深圳蓝因机器人科技有限公司 | Point cloud data processing method and device |
CN113763308A (en) * | 2020-09-23 | 2021-12-07 | 北京京东乾石科技有限公司 | Ground detection method, device, server and medium |
CN114581361A (en) * | 2021-06-28 | 2022-06-03 | 广州极飞科技股份有限公司 | Object form measuring method, device, equipment and storage medium |
CN114842450A (en) * | 2022-05-11 | 2022-08-02 | 合众新能源汽车有限公司 | Driving region detection method, device and equipment |
CN115712298A (en) * | 2022-10-25 | 2023-02-24 | 太原理工大学 | Autonomous navigation method for robot running in channel based on single-line laser radar |
CN117710244A (en) * | 2024-02-05 | 2024-03-15 | 湖南裕工新能科技有限公司 | Vehicle-mounted assembly material alignment intelligent detection method and system |
CN117710244B (en) * | 2024-02-05 | 2024-04-26 | 湖南裕工新能科技有限公司 | Vehicle-mounted assembly material alignment intelligent detection method and system |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106530380A (en) * | 2016-09-20 | 2017-03-22 | 长安大学 | Ground point cloud segmentation method based on three-dimensional laser radar |
CN106842231A (en) * | 2016-11-08 | 2017-06-13 | 长安大学 | A kind of road edge identification and tracking |
CN106937910A (en) * | 2017-03-20 | 2017-07-11 | 杭州视氪科技有限公司 | A kind of barrier and ramp detecting system and method |
WO2018087703A1 (en) * | 2016-11-11 | 2018-05-17 | Info Solution S.P.A. | Method and device for driving a self-moving vehicle and related driving system |
CN109074490A (en) * | 2018-07-06 | 2018-12-21 | 深圳前海达闼云端智能科技有限公司 | Path detection method, related device and computer readable storage medium |
CN109407115A (en) * | 2018-12-25 | 2019-03-01 | 中山大学 | A kind of road surface extraction system and its extracting method based on laser radar |
CN110008941A (en) * | 2019-06-05 | 2019-07-12 | 长沙智能驾驶研究院有限公司 | Drivable region detection method, device, computer equipment and storage medium |
CN110084116A (en) * | 2019-03-22 | 2019-08-02 | 深圳市速腾聚创科技有限公司 | Pavement detection method, apparatus, computer equipment and storage medium |
US20190324148A1 (en) * | 2018-04-19 | 2019-10-24 | Faraday&Future Inc. | System and method for ground and free-space detection |
CN110425977A (en) * | 2019-07-10 | 2019-11-08 | 上海工程技术大学 | A kind of conductor rail geometric parameter detection system and detection method |
-
2020
- 2020-01-09 CN CN202010023358.1A patent/CN111208533A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106530380A (en) * | 2016-09-20 | 2017-03-22 | 长安大学 | Ground point cloud segmentation method based on three-dimensional laser radar |
CN106842231A (en) * | 2016-11-08 | 2017-06-13 | 长安大学 | A kind of road edge identification and tracking |
WO2018087703A1 (en) * | 2016-11-11 | 2018-05-17 | Info Solution S.P.A. | Method and device for driving a self-moving vehicle and related driving system |
CN106937910A (en) * | 2017-03-20 | 2017-07-11 | 杭州视氪科技有限公司 | A kind of barrier and ramp detecting system and method |
US20190324148A1 (en) * | 2018-04-19 | 2019-10-24 | Faraday&Future Inc. | System and method for ground and free-space detection |
CN109074490A (en) * | 2018-07-06 | 2018-12-21 | 深圳前海达闼云端智能科技有限公司 | Path detection method, related device and computer readable storage medium |
CN109407115A (en) * | 2018-12-25 | 2019-03-01 | 中山大学 | A kind of road surface extraction system and its extracting method based on laser radar |
CN110084116A (en) * | 2019-03-22 | 2019-08-02 | 深圳市速腾聚创科技有限公司 | Pavement detection method, apparatus, computer equipment and storage medium |
CN110008941A (en) * | 2019-06-05 | 2019-07-12 | 长沙智能驾驶研究院有限公司 | Drivable region detection method, device, computer equipment and storage medium |
CN110425977A (en) * | 2019-07-10 | 2019-11-08 | 上海工程技术大学 | A kind of conductor rail geometric parameter detection system and detection method |
Non-Patent Citations (1)
Title |
---|
CHEN TONGTONG等: ""3D lidar-based ground segmentation"", 《THE FIRST ASIAN CONFERENCE ON PATTERN RECOGNITION》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112036274A (en) * | 2020-08-19 | 2020-12-04 | 江苏智能网联汽车创新中心有限公司 | Driving region detection method and device, electronic equipment and storage medium |
CN113763308A (en) * | 2020-09-23 | 2021-12-07 | 北京京东乾石科技有限公司 | Ground detection method, device, server and medium |
CN113763308B (en) * | 2020-09-23 | 2024-04-05 | 北京京东乾石科技有限公司 | Ground detection method, device, server and medium |
CN112180343A (en) * | 2020-09-30 | 2021-01-05 | 东软睿驰汽车技术(沈阳)有限公司 | Laser point cloud data processing method, device and equipment and unmanned system |
CN113050106A (en) * | 2021-02-05 | 2021-06-29 | 上海擎朗智能科技有限公司 | Ground detection method, device, electronic equipment, system and medium |
CN114581361B (en) * | 2021-06-28 | 2023-03-07 | 广州极飞科技股份有限公司 | Object form measuring method, device, equipment and storage medium |
CN114581361A (en) * | 2021-06-28 | 2022-06-03 | 广州极飞科技股份有限公司 | Object form measuring method, device, equipment and storage medium |
CN113689329A (en) * | 2021-07-02 | 2021-11-23 | 上海工程技术大学 | Shortest path interpolation method for enhancing sparse point cloud |
CN113744323A (en) * | 2021-08-11 | 2021-12-03 | 深圳蓝因机器人科技有限公司 | Point cloud data processing method and device |
CN113744323B (en) * | 2021-08-11 | 2023-12-19 | 深圳蓝因机器人科技有限公司 | Point cloud data processing method and device |
CN114842450A (en) * | 2022-05-11 | 2022-08-02 | 合众新能源汽车有限公司 | Driving region detection method, device and equipment |
WO2023216470A1 (en) * | 2022-05-11 | 2023-11-16 | 合众新能源汽车股份有限公司 | Drivable area detection method and apparatus, and device |
CN115712298A (en) * | 2022-10-25 | 2023-02-24 | 太原理工大学 | Autonomous navigation method for robot running in channel based on single-line laser radar |
CN117710244A (en) * | 2024-02-05 | 2024-03-15 | 湖南裕工新能科技有限公司 | Vehicle-mounted assembly material alignment intelligent detection method and system |
CN117710244B (en) * | 2024-02-05 | 2024-04-26 | 湖南裕工新能科技有限公司 | Vehicle-mounted assembly material alignment intelligent detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111208533A (en) | Real-time ground detection method based on laser radar | |
CN109858460B (en) | Lane line detection method based on three-dimensional laser radar | |
CN106650640B (en) | Negative obstacle detection method based on laser radar point cloud local structure characteristics | |
CN107045629B (en) | Multi-lane line detection method | |
CN106022381B (en) | Automatic extraction method of street lamp pole based on vehicle-mounted laser scanning point cloud | |
CN113156421A (en) | Obstacle detection method based on information fusion of millimeter wave radar and camera | |
CN111079611B (en) | Automatic extraction method for road surface and marking line thereof | |
CN110781827A (en) | Road edge detection system and method based on laser radar and fan-shaped space division | |
JP5867596B2 (en) | Three-dimensional object detection apparatus and three-dimensional object detection method | |
CN108828621A (en) | Obstacle detection and road surface partitioning algorithm based on three-dimensional laser radar | |
GB2532948A (en) | Objection recognition in a 3D scene | |
CN111325138B (en) | Road boundary real-time detection method based on point cloud local concave-convex characteristics | |
CN108229342B (en) | Automatic sea surface ship target detection method | |
CN113920134B (en) | Slope ground point cloud segmentation method and system based on multi-line laser radar | |
CN109948635B (en) | Target identification method and device based on laser scanning | |
CN107895151A (en) | Method for detecting lane lines based on machine vision under a kind of high light conditions | |
CN106504289B (en) | indoor target detection method and device | |
Lin et al. | Construction of fisheye lens inverse perspective mapping model and its applications of obstacle detection | |
CN112597839B (en) | Road boundary detection method based on vehicle-mounted millimeter wave radar | |
CN110532963B (en) | Vehicle-mounted laser radar point cloud driven road marking accurate extraction method | |
CN114764871B (en) | Urban building attribute extraction method based on airborne laser point cloud | |
CN114782729A (en) | Real-time target detection method based on laser radar and vision fusion | |
CN110659552A (en) | Tramcar obstacle detection and alarm method | |
CN116109601A (en) | Real-time target detection method based on three-dimensional laser radar point cloud | |
Baluyan et al. | Novel approach for rooftop detection using support vector machine |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200529 |