CN116071550B - Laser radar dust point cloud filtering method - Google Patents
Laser radar dust point cloud filtering method Download PDFInfo
- Publication number
- CN116071550B CN116071550B CN202310135126.9A CN202310135126A CN116071550B CN 116071550 B CN116071550 B CN 116071550B CN 202310135126 A CN202310135126 A CN 202310135126A CN 116071550 B CN116071550 B CN 116071550B
- Authority
- CN
- China
- Prior art keywords
- point
- dust
- points
- point cloud
- laser radar
- 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
- 239000000428 dust Substances 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 15
- 238000001914 filtration Methods 0.000 title claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 18
- 230000035515 penetration Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 4
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000002834 transmittance Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 2
- 230000007547 defect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/273—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion removing elements interfering with the pattern to be recognised
-
- 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
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The application discloses a laser radar dust point cloud filtering method, which comprises the steps of acquiring point cloud data and space coordinate information based on a laser radar, and calculating data characteristics of the point cloud; distinguishing ground points from non-ground points under polar coordinates, and selecting and removing free points; adjusting the threshold value of the subsequent screening point based on the distance between the point and the laser radar; the method comprises the steps of (1) pre-screening free points for characteristics of adjacent points in the same row, clustering the extracted free points based on breadth-first search, and performing characteristic calculation on dust point clouds after clustering based on PCA; and calculating the penetration value of the cluster point cloud, and performing counter selection on the characteristics of the special target to obtain the finally confirmed dust characteristics. According to the application, by acquiring the power supply data and adopting a multi-key cloud data processing mode, whether the dust point is determined, and finally, the interference of the dust point on obstacle detection is eliminated.
Description
Technical Field
The application relates to the technical field of automatic driving, in particular to a laser radar dust point cloud filtering method.
Background
Autopilot, as a new travel mode, gradually merges into people's daily life, and has excellent performance in various fields. Such as in the working environment of open-pit mines, can also help people to perform work production. However, a large amount of dust floats in the air in the open-pit mine, and a large amount of dust has a great test on the recognition obstacle avoidance of the automatic driving sensor.
The prior art has the defects that at present, in an open-air mine, when a large amount of dust is blown, the dust can cause obstacles to be formed in laser radar point clouds, and the traditional algorithm cannot distinguish the point clouds formed by the dust from the point clouds of real obstacles, so that false alarm is caused.
Disclosure of Invention
The application aims to overcome the defects in the prior art, and aims to achieve the purposes, a laser radar dust point cloud filtering method is adopted to solve the problems in the background art.
A laser radar dust point cloud filtering method specifically comprises the following steps:
step S1, acquiring point cloud data and space coordinate information based on a laser radar, and calculating data characteristics of the point cloud;
s2, distinguishing ground points from non-ground points under polar coordinates, and selecting and removing free points;
step S3, adjusting the threshold value of the subsequent screening point based on the distance between the point and the laser radar;
s4, pre-screening free points for the characteristics of adjacent points in the same row, clustering the extracted free points based on breadth-first search, and performing characteristic calculation on the clustered dust point cloud based on PCA;
and S5, calculating the penetration value of the cluster point cloud, and performing counter selection on the characteristics of the special target to obtain the finally confirmed dust characteristics.
As a further aspect of the application: the specific steps in the step S1 include:
s11, acquiring point cloud data of objects in the surrounding environment based on a laser radar, and calculating rows and columns of points in the point cloud data and other characteristics;
calculating an incidence angle in a vertical direction by an incidence angle formula, wherein the incidence angle formula is as follows:
α＝arctan(z,(x 2 +y 2 ) 0.5 )；
confirming the number of lines of the point according to the obtained incidence angle and marking;
calculating a horizontal offset angle through an offset angle formula, wherein the offset angle formula is as follows:
β＝arctan(x,y)；
confirming the column number of the point according to the obtained offset angle and marking;
s12, calculating the distance characteristic according to the distance characteristic of the cloud point, wherein the formula is as follows:
d＝(x 2 +y 2 ) 0.5 ；
calculating the distance variance of N points before and after each point as a variance characteristic and the smoothness of N points before and after each point, wherein the smoothness is the sum of the differences between the N points before and after and the current point distance characteristic;
calculating the curvatures of N points before and after each point, calculating standard forward module length through the smoothness of x and y of the N points before and after each point, and normalizing the smoothness, wherein the curvatures of x and y in the directions are products of the smoothness after positive and negative normalization respectively;
and then calculating the curvature of the current point according to a curvature formula, wherein the formula is as follows:
k＝(cur_x 2 +cur_y 2 ) 0.5 ；
wherein curv_x and curv_y are curvatures in x and y directions, respectively.
As a further aspect of the application: the specific steps in the step S3 include:
obtaining the distance between each point in the point cloud and the laser radar;
by adopting a sixth-order polynomial to perform threshold adjustment calculation, the situation that the threshold value of each point is larger as the distance from the laser radar is larger is avoided, dust is difficult to identify, and the point cloud is sparse and is erroneously identified as dust.
As a further aspect of the application: the specific step of feature preliminary screening free points of the same row of adjacent points in the step S4 comprises the following steps:
the difference value of the distance characteristic between the current point and the previous point is larger than a threshold value, or the variance characteristic is larger than the threshold value or the smoothness characteristic is larger than the threshold value or the curvature characteristic is larger than the threshold value;
if the difference value of the distance characteristic between the M points and the point after the start of the point is smaller than the threshold value, the point is not considered as a free point, but is a starting point of another obstacle point cloud;
wherein, if a point meets the primary screening condition and does not meet the obstacle starting point condition, the point is considered as a dust starting point; meanwhile, the corresponding subsequent P points are directly regarded as dust points and enter the next analysis;
and (3) independently storing all dust points as a primary screening dust set to enter the next step of processing, and directly storing other unselected point clouds into a point cloud set to be output.
As a further aspect of the application: in the step S4, the specific steps of clustering the extracted free points based on breadth-first search and performing feature calculation on the clustered dust point cloud based on PCA include:
circulating all dust points, carrying out feature calculation on 4 points which are arranged in the same column of each dust adjacent row and the same row of each adjacent column, and clustering;
and after the clustering is completed, carrying out principal component analysis on the point clouds of each category, and confirming the characteristic values corresponding to the first principal component and the second principal component.
As a further aspect of the application: the specific steps in the step S5 include:
calculating penetration values in the same row according to the clustered point cloud data;
meanwhile, because of the existence of special obstacles, the point cloud of the special obstacles has little extension in the transverse direction and is mistaken as dust points, and the situation that the slender obstacles are confirmed as dust is eliminated through searching in the vertical direction;
and finally, storing the dust-removing labels which are subjected to the analysis according to the clustered principal components and have the first principal component or the second principal component which are larger than a threshold value and the transmittance which is smaller than the threshold value or meet the characteristics of the slender obstacle into a point cloud set to be output, and outputting a result.
Compared with the prior art, the application has the following technical effects:
by adopting the technical scheme, the point cloud data is acquired based on the laser radar, the characteristics of the point cloud data are calculated, the ground points are removed, screening is carried out, and irregular point cloud data are acquired by primarily screening free points; and then BFS is utilized to carry out three-dimensional characteristic analysis on the clustered dust points, and whether the dust points are dust points is confirmed through comprehensive calculation of the penetration rate and PCA. Finally, the interference of dust points on obstacle detection is eliminated. The point cloud formed by dust and the point cloud of a real obstacle are distinguished, and false alarm is caused.
Drawings
The following detailed description of specific embodiments of the application refers to the accompanying drawings, in which:
fig. 1 is a schematic step diagram of a dust filtration method according to some embodiments of the present disclosure.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, in an embodiment of the present application, a method for filtering dust point cloud of a laser radar includes the following specific steps:
step S1, acquiring point cloud data and space coordinate information based on a laser radar, and calculating data characteristics of the point cloud, wherein the specific steps comprise:
s11, acquiring point cloud data of objects in the surrounding environment based on a laser radar, and calculating rows and columns of points in the point cloud data and other characteristics;
calculating an incidence angle in a vertical direction by an incidence angle formula, wherein the incidence angle formula is as follows:
α＝arctan(z,(x 2 +y 2 ) 0.5 )；
confirming the number of lines of the point according to the obtained incidence angle and marking;
calculating a horizontal offset angle through an offset angle formula, wherein the offset angle formula is as follows:
β＝arctan(x,y)；
confirming the column number of the point according to the obtained offset angle and marking;
s12, calculating the distance characteristic according to the distance characteristic of the cloud point, wherein the formula is as follows:
d＝(x 2 +y 2 ) 0.5 ；
calculating the distance variance of N points before and after each point as a variance characteristic and the smoothness of N points before and after each point, wherein the smoothness is the sum of the differences between the N points before and after and the current point distance characteristic;
calculating the curvatures of N points before and after each point, calculating standard forward module length through the smoothness of x and y of the N points before and after each point, and normalizing the smoothness, wherein the curvatures of x and y in the directions are products of the smoothness after positive and negative normalization respectively;
and then calculating the curvature of the current point according to a curvature formula, wherein the formula is as follows:
k＝(cur_x 2 +cur_y 2 ) 0.5 ；
wherein curv_x and curv_y are curvatures in x and y directions, respectively.
S2, distinguishing ground points from non-ground points under polar coordinates, and selecting and removing free points;
in this embodiment, the ground points and the non-ground points are distinguished and the free points are selected and removed by the existing method under the polar coordinates.
Step S3, adjusting the threshold value of the subsequent screening point based on the distance between the point and the laser radar, wherein the specific steps comprise:
obtaining the distance between each point in the point cloud and the laser radar;
by adopting a sixth-order polynomial to perform threshold adjustment calculation, the situation that the threshold value of each point is larger as the distance from the laser radar is larger is avoided, dust is difficult to identify, and the point cloud is sparse and is erroneously identified as dust.
In this embodiment, specifically, the farther the point is from the lidar, the larger the threshold thereof is. I.e. more difficult to identify as dust, to avoid that distant obstacles are mistakenly identified as dust because of sparse point clouds. The threshold adjustment calculation is performed using a sixth order polynomial.
S4, pre-screening free points for the characteristics of adjacent points in the same row, clustering the extracted free points based on breadth-first search, and performing characteristic calculation on the clustered dust point cloud based on PCA;
in step S41, in this embodiment, the specific steps of feature preliminary screening free points of adjacent points in the same row include:
first, preliminary screening is carried out, and conditions of free points are set:
the difference value of the distance characteristic between the current point and the previous point is larger than a threshold value, or the variance characteristic is larger than the threshold value or the smoothness characteristic is larger than the threshold value or the curvature characteristic is larger than the threshold value;
if the difference value of the distance characteristic between the M points and the point after the start of the point is smaller than the threshold value, the point is not considered as a free point, but is a starting point of another obstacle point cloud;
wherein, if a point meets the primary screening condition and does not meet the obstacle starting point condition, the point is considered as a dust starting point; meanwhile, the corresponding subsequent P points are directly regarded as dust points and enter the next analysis;
and (3) independently storing all dust points as a primary screening dust set to enter the next step of processing, and directly storing other unselected point clouds into a point cloud set to be output.
Step S42, clustering the extracted free points based on breadth-first search, and performing feature calculation on the clustered dust point cloud based on PCA, wherein the specific steps comprise:
first, a cyclic calculation is performed for each dust point:
circulating all dust points, carrying out feature calculation on 4 points which are arranged in the same column of each dust adjacent row and the same row of each adjacent column, and clustering;
in this embodiment, taking a point as an example, the distance characteristic between the adjacent point and the current dust point is taken, where l is long and s is short. The included angle between the two points is the angular resolution alpha of the laser radar, wherein the vertical angular resolution of the laser radar is taken as if the rows are different. And taking the horizontal angle resolution of the laser radar as the same row and different columns.
Calculation is performed according to the formula arctan (s sin (α), (l-s cos) (α));
if the angle is larger than the threshold value, the adjacent point and the current point are considered to belong to the same category, the adjacent point is taken as a central point to continue searching in other three directions until all dust points are circulated, and finally clustering is completed.
Step S43, performing feature calculation on the clustered dust point cloud based on PCA:
and after the clustering is completed, carrying out principal component analysis on the point clouds of each category, and confirming the characteristic values corresponding to the first principal component and the second principal component.
S5, calculating the penetration value of the cluster point cloud, and performing back selection on the characteristics of the special target to obtain the finally confirmed dust characteristics, wherein the method comprises the following specific steps of:
step S51, calculating penetration values of the cluster point clouds:
calculating penetration values in the same row according to the clustered point cloud data;
in this embodiment, if the number of rows of dust spots in the same row is 3, 9, 11, the calculated penetration is ((9-3) + (11-9))/3= 2.66667.
Step S52, feature counter selection of special targets:
meanwhile, because of the existence of special obstacles, the point cloud of the special obstacles has little extension in the transverse direction and is mistaken as dust points, and the situation that the slender obstacles are confirmed as dust is eliminated through searching in the vertical direction;
and step S53, finally, storing dust removal labels meeting the characteristics of the elongated obstacles according to the fact that the first principal component or the second principal component analyzed by the principal components after clustering is larger than a threshold value, the transmittance is smaller than the threshold value or the characteristics of the elongated obstacles, and outputting a result.
Although embodiments of the present application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.
Claims (5)
1. A laser radar dust point cloud filtering method is characterized by comprising the following specific steps:
step S1, acquiring point cloud data and space coordinate information based on a laser radar, and calculating data characteristics of the point cloud, wherein the specific steps comprise:
s11, acquiring point cloud data of objects in the surrounding environment based on a laser radar, and calculating rows and columns of points in the point cloud data and other characteristics;
calculating an incidence angle in a vertical direction by an incidence angle formula, wherein the incidence angle formula is as follows:
；
confirming the number of lines of the point according to the obtained incidence angle and marking;
calculating a horizontal offset angle through an offset angle formula, wherein the offset angle formula is as follows:
；
confirming the column number of the point according to the obtained offset angle and marking;
s12, calculating the distance characteristic according to the distance characteristic of the cloud point, wherein the formula is as follows:
；
calculating the distance variance of N points before and after each point as a variance characteristic and the smoothness of N points before and after each point, wherein the smoothness is the sum of the differences between the N points before and after and the current point distance characteristic;
calculating the curvatures of N points before and after each point, calculating standard forward module length through the smoothness of x and y of the N points before and after each point, and normalizing the smoothness, wherein the curvatures of x and y in the directions are products of the smoothness after positive and negative normalization respectively;
and then calculating the curvature of the current point according to a curvature formula, wherein the formula is as follows:
；
wherein , and />Curvature in x and y directions, respectively;
s2, distinguishing ground points from non-ground points under polar coordinates, and selecting and removing free points;
step S3, adjusting the threshold value of the subsequent screening point based on the distance between the point and the laser radar;
s4, pre-screening free points for the characteristics of adjacent points in the same row, clustering the extracted free points based on breadth-first search, and performing characteristic calculation on the clustered dust point cloud based on PCA;
and S5, calculating the penetration value of the cluster point cloud, and performing counter selection on the characteristics of the special target to obtain the finally confirmed dust characteristics.
2. The method for filtering the dust point cloud of the lidar according to claim 1, wherein the specific steps in the step S3 include:
obtaining the distance between each point in the point cloud and the laser radar;
by adopting a sixth-order polynomial to perform threshold adjustment calculation, the situation that the threshold value of each point is larger as the distance from the laser radar is larger is avoided, dust is difficult to identify, and the point cloud is sparse and is erroneously identified as dust.
3. The method for filtering the laser radar dust point cloud according to claim 1, wherein the specific step of pre-screening the free points for the features of the adjacent points in the same row in step S4 includes:
the difference value of the distance characteristic between the current point and the previous point is larger than a threshold value, or the variance characteristic is larger than the threshold value or the smoothness characteristic is larger than the threshold value or the curvature characteristic is larger than the threshold value;
if the difference value of the distance characteristic between the M points and the point after the start of the point is smaller than the threshold value, the point is not considered as a free point, but is a starting point of another obstacle point cloud;
wherein, if a point meets the primary screening condition and does not meet the obstacle starting point condition, the point is considered as a dust starting point; meanwhile, the corresponding subsequent P points are directly regarded as dust points and enter the next analysis;
and (3) independently storing all dust points as a primary screening dust set to enter the next step of processing, and directly storing other unselected point clouds into a point cloud set to be output.
4. The method for filtering the dust point cloud of the laser radar according to claim 1, wherein the specific steps of clustering the extracted free points based on breadth-first search and performing feature calculation on the clustered dust point cloud based on PCA in the step S4 include:
circulating all dust points, carrying out feature calculation on 4 points which are arranged in the same column of each dust adjacent row and the same row of each adjacent column, and clustering;
and after the clustering is completed, carrying out principal component analysis on the point clouds of each category, and confirming the characteristic values corresponding to the first principal component and the second principal component.
5. The method for filtering laser radar dust point clouds according to claim 1, wherein the specific steps in the step S5 include:
calculating penetration values in the same row according to the clustered point cloud data;
meanwhile, because of the existence of special obstacles, the point cloud of the special obstacles has little extension in the transverse direction and is mistaken as dust points, and the situation that the slender obstacles are confirmed as dust is eliminated through searching in the vertical direction;
and finally, storing the dust-removing labels which are subjected to the analysis according to the clustered principal components and have the first principal component or the second principal component which are larger than a threshold value and the transmittance which is smaller than the threshold value or meet the characteristics of the slender obstacle into a point cloud set to be output, and outputting a result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310135126.9A CN116071550B (en) | 2023-02-09 | 2023-02-09 | Laser radar dust point cloud filtering method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310135126.9A CN116071550B (en) | 2023-02-09 | 2023-02-09 | Laser radar dust point cloud filtering method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116071550A CN116071550A (en) | 2023-05-05 |
| CN116071550B true CN116071550B (en) | 2023-10-20 |
Family
ID=86179959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310135126.9A Active CN116071550B (en) | 2023-02-09 | 2023-02-09 | Laser radar dust point cloud filtering method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116071550B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106408581A (en) * | 2016-09-12 | 2017-02-15 | 厦门大学 | Rapid three-dimensional point cloud linear extraction method |
| CN109961440A (en) * | 2019-03-11 | 2019-07-02 | 重庆邮电大学 | A kind of three-dimensional laser radar point cloud Target Segmentation method based on depth map |
| CN111880196A (en) * | 2020-06-29 | 2020-11-03 | 安徽海博智能科技有限责任公司 | Unmanned mine car anti-interference method, system and computer equipment |
| CN111929699A (en) * | 2020-07-21 | 2020-11-13 | 北京建筑大学 | Laser radar inertial navigation odometer considering dynamic obstacles and mapping method and system |
| CN112099046A (en) * | 2020-09-16 | 2020-12-18 | 辽宁工程技术大学 | Airborne LIDAR three-dimensional plane detection method based on multi-valued voxel model |
| CN112258600A (en) * | 2020-10-19 | 2021-01-22 | 浙江大学 | Simultaneous positioning and map construction method based on vision and laser radar |
| CN112666557A (en) * | 2020-11-20 | 2021-04-16 | 芜湖格陆博智能科技有限公司 | Method for acquiring mining area information based on laser radar and millimeter wave radar |
| CN113138396A (en) * | 2021-06-21 | 2021-07-20 | 上海伯镭智能科技有限公司 | Dust and obstacle detection method based on laser radar |
| CN113330332A (en) * | 2021-04-29 | 2021-08-31 | 华为技术有限公司 | Laser detection device, control method and control device thereof, and terminal |
| CN114355381A (en) * | 2021-12-31 | 2022-04-15 | 安徽海博智能科技有限责任公司 | Laser radar point cloud quality detection and improvement method |
| CN114692734A (en) * | 2022-03-11 | 2022-07-01 | 三一智矿科技有限公司 | Mining area dust filtering method, device, equipment and medium based on point cloud |
| CN114994684A (en) * | 2022-06-01 | 2022-09-02 | 湖南大学无锡智能控制研究院 | Multi-radar data fusion obstacle detection method and system in dust raising scene |
| CN115131417A (en) * | 2022-07-19 | 2022-09-30 | 南开大学 | Laser point cloud 2D-3D bimodal interaction enhancement irregular wire detection method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11449061B2 (en) * | 2016-02-29 | 2022-09-20 | AI Incorporated | Obstacle recognition method for autonomous robots |
| CN114296093A (en) * | 2016-11-29 | 2022-04-08 | 布莱克莫尔传感器和分析有限责任公司 | Method and system for classifying objects in a point cloud data set |
-
2023
- 2023-02-09 CN CN202310135126.9A patent/CN116071550B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106408581A (en) * | 2016-09-12 | 2017-02-15 | 厦门大学 | Rapid three-dimensional point cloud linear extraction method |
| CN109961440A (en) * | 2019-03-11 | 2019-07-02 | 重庆邮电大学 | A kind of three-dimensional laser radar point cloud Target Segmentation method based on depth map |
| CN111880196A (en) * | 2020-06-29 | 2020-11-03 | 安徽海博智能科技有限责任公司 | Unmanned mine car anti-interference method, system and computer equipment |
| CN111929699A (en) * | 2020-07-21 | 2020-11-13 | 北京建筑大学 | Laser radar inertial navigation odometer considering dynamic obstacles and mapping method and system |
| CN112099046A (en) * | 2020-09-16 | 2020-12-18 | 辽宁工程技术大学 | Airborne LIDAR three-dimensional plane detection method based on multi-valued voxel model |
| CN112258600A (en) * | 2020-10-19 | 2021-01-22 | 浙江大学 | Simultaneous positioning and map construction method based on vision and laser radar |
| CN112666557A (en) * | 2020-11-20 | 2021-04-16 | 芜湖格陆博智能科技有限公司 | Method for acquiring mining area information based on laser radar and millimeter wave radar |
| CN113330332A (en) * | 2021-04-29 | 2021-08-31 | 华为技术有限公司 | Laser detection device, control method and control device thereof, and terminal |
| CN113138396A (en) * | 2021-06-21 | 2021-07-20 | 上海伯镭智能科技有限公司 | Dust and obstacle detection method based on laser radar |
| CN114355381A (en) * | 2021-12-31 | 2022-04-15 | 安徽海博智能科技有限责任公司 | Laser radar point cloud quality detection and improvement method |
| CN114692734A (en) * | 2022-03-11 | 2022-07-01 | 三一智矿科技有限公司 | Mining area dust filtering method, device, equipment and medium based on point cloud |
| CN114994684A (en) * | 2022-06-01 | 2022-09-02 | 湖南大学无锡智能控制研究院 | Multi-radar data fusion obstacle detection method and system in dust raising scene |
| CN115131417A (en) * | 2022-07-19 | 2022-09-30 | 南开大学 | Laser point cloud 2D-3D bimodal interaction enhancement irregular wire detection method |
Non-Patent Citations (5)
| Title |
|---|
| Analyzing point cloud of coal mining process in much dust environment based on dynamic graph convolution neural network;Zhizhong Xing 等;《Environmental Science and Pollution Research》;4044–4061 * |
| Dust Removal from 3D Point Cloud Data in Mine Plane Areas Based on Orthogonal Total Least Squares Fitting and GA-TELM;Jingli Wang 等;《Computational Intelligence and Neuroscience》;1-8 * |
| 农田环境下基于激光点云的玉米配准研究及果穗识别;谢良毅;《中国优秀硕士学位论文全文数据库 基础科学辑》;A005-391 * |
| 基于三维激光扫描技术的爆堆特征信息提取与应用;刘士彪;《中国优秀硕士学位论文全文数据库 工程科技I辑》;B021-183 * |
| 基于激光雷达扫描和视觉识别结合的水面环境探测;张堡瑞;《中国优秀硕士学位论文全文数据库 工程科技I辑》;B027-1731 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116071550A (en) | 2023-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109100741B (en) | Target detection method based on 3D laser radar and image data | |
| CN110781827B (en) | Road edge detection system and method based on laser radar and fan-shaped space division | |
| CN108830199B (en) | Method and device for identifying traffic light signal, readable medium and electronic equipment | |
| US6859555B1 (en) | Fast dominant circle detection through horizontal and vertical scanning | |
| CN107665351B (en) | Airport detection method based on difficult sample mining | |
| US6327386B1 (en) | Key character extraction and lexicon reduction for cursive text recognition | |
| CN111444769B (en) | Laser radar human leg detection method based on multi-scale self-adaptive random forest | |
| CN110349260B (en) | Automatic pavement marking extraction method and device | |
| CN105608441B (en) | Vehicle type recognition method and system | |
| CN113936198B (en) | Low-beam laser radar and camera fusion method, storage medium and device | |
| CN107563370B (en) | Visual attention mechanism-based marine infrared target detection method | |
| CN111652857A (en) | Infrared detection method for insulator defects | |
| CN114488194A (en) | Method for detecting and identifying targets under structured road of intelligent driving vehicle | |
| Bogoslavskyi et al. | Analyzing the quality of matched 3D point clouds of objects | |
| CN111208484B (en) | Main lobe dense false target removing method based on angle information | |
| CN113740874A (en) | Road edge detection method, electronic equipment and vehicle | |
| CN116071550B (en) | Laser radar dust point cloud filtering method | |
| CN111832497B (en) | Text detection post-processing method based on geometric features | |
| CN110335282B (en) | Contour line segment feature extraction method based on grids | |
| JP2015232805A (en) | Image processing method, image processor, and image processing program | |
| CN116385801A (en) | Tree classification method based on multispectral deep learning and point cloud technology | |
| US20220171061A1 (en) | Apparatus and method for recognizing high-elevation structure using lidar sensor | |
| Najdataei et al. | Lisco: A continuous approach in lidar point-cloud clustering | |
| Sun et al. | Object Detection with Geometrical Context Feedback Loop. | |
| CN112651278A (en) | Detection data processing method based on feature screening |
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 |