CN116482700A - Robust unmanned plane laser radar line-imitating tower recognition method and device - Google Patents

Robust unmanned plane laser radar line-imitating tower recognition method and device Download PDF

Info

Publication number
CN116482700A
CN116482700A CN202310441901.3A CN202310441901A CN116482700A CN 116482700 A CN116482700 A CN 116482700A CN 202310441901 A CN202310441901 A CN 202310441901A CN 116482700 A CN116482700 A CN 116482700A
Authority
CN
China
Prior art keywords
point cloud
tower
power line
seed
current frame
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
Application number
CN202310441901.3A
Other languages
Chinese (zh)
Inventor
张勇
吴芳芳
郭彦明
孙喜亮
张衡
陈琳海
高上
徐光彩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Lvtuzhi New Technology Co ltd
Beijing Digital Green Earth Technology Co ltd
Original Assignee
Shenzhen Lvtuzhi New Technology Co ltd
Beijing Digital Green Earth Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Lvtuzhi New Technology Co ltd, Beijing Digital Green Earth Technology Co ltd filed Critical Shenzhen Lvtuzhi New Technology Co ltd
Priority to CN202310441901.3A priority Critical patent/CN116482700A/en
Publication of CN116482700A publication Critical patent/CN116482700A/en
Pending legal-status Critical Current

Links

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
    • G01S17/42—Simultaneous measurement of distance and other co-ordinates
    • 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
    • G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
    • 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
    • G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
    • G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
    • G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
    • G01S19/42—Determining position
    • G01S19/45—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement
    • G01S19/46—Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement the supplementary measurement being of a radio-wave signal type
    • 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)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Optical Radar Systems And Details Thereof (AREA)

Abstract

The invention discloses a robust unmanned aerial vehicle laser radar line-imitating tower recognition method and device, wherein unmanned aerial vehicle laser radar is utilized to scan a transmission line scene to obtain line and tower laser point clouds; firstly, extracting a power line channel and a pole tower of each frame of data in the flight process, and filtering the extracted pole tower point cloud in different ranges by utilizing a power line linear equation of the current channel to obtain pole tower seed points and alternative points; then, respectively accumulating, sampling and clustering the seed points of the tower obtained by filtering and the alternative point cloud frame by frame, judging whether the clustered seed points reach the threshold value of the seed points, and further determining whether the tower exists; and finally, restoring the shape of the tower point cloud by utilizing the overlapping relation of the seed points and the alternative points, thereby identifying the tower, and finally calculating the coordinates of the tower and the central height of the tower.

Description

Robust unmanned plane laser radar line-imitating tower recognition method and device
Technical Field
The invention relates to the field of unmanned aerial vehicle laser radar line-imitating flight, in particular to a robust unmanned aerial vehicle laser radar line-imitating tower identification method and device.
Background
Utilize unmanned aerial vehicle laser radar system imitative line flight automatic inspection process, unmanned aerial vehicle's flight position and gesture probably lead to the laser to the condition of shaft tower scanning insufficiency in certain time quantum, and accidental flight rocks and can lead to partial shaft tower data loss, or unmanned aerial vehicle is being close to the in-process of shaft tower gradually, and shaft tower data is complete gradually, and not can carry out the overall view to the shaft tower immediately.
At present, the main technical defects of the main line-imitating flying are as follows: 1. the scanning range of the laser is wider, and a plurality of lines can exist below, so that a plurality of towers can be observed at the same time, and at the moment, the currently observed towers need to be distinguished into which line in the plurality of lines belongs; 2. it is also a problem to be solved how to effectively distinguish unrelated towers or ground objects by observing a column-like ground object like a tower. Therefore, a method is needed to keep comprehensive processing on multi-frame observation data, quickly and continuously and stably identify towers, and the earlier the existence of a front tower is identified, the more favorable the unmanned aerial vehicle line-simulating system to make corresponding decisions. Otherwise, because of the condition that the integrity of single frame data is insufficient or other interference lines exist to cause the recognition error of the towers, the condition is sometimes present, which is unfavorable for making flight decisions. 3. When the unmanned aerial vehicle flies in a line-imitating mode, the laser radar can scan a flown tower, the flown tower is recorded in historical tower information, and filtering is needed for the flown tower.
Disclosure of Invention
The invention aims to provide a robust unmanned aerial vehicle laser radar line-imitating tower recognition method and device, which solve the technical problems pointed out in the prior art.
The invention provides a robust unmanned aerial vehicle laser radar line-imitating tower recognition device, which comprises a line-imitating flying unmanned aerial vehicle;
the line-imitating flying unmanned aerial vehicle comprises a laser radar module, a central processing unit and a GPS positioning module;
the central processing unit is respectively connected with the laser radar module and the GPS positioning module;
the laser radar module is used for acquiring all power line channel data of the current frame and the pole tower point cloud of the current frame in real time;
the current frame power line channel data comprises line linear equations of all power line channels of the current frame; the pole tower point cloud comprises coordinates of the pole tower point cloud and the height of the pole tower point cloud;
the GPS positioning module is used for acquiring the coordinate position of the line-simulated flying unmanned aerial vehicle;
the central processing unit is used for acquiring all power line channel data of the current frame and all tower point clouds of the current frame in real time according to the laser radar module, and acquiring the tower point clouds of the current frame from all the tower point clouds of the current frame; screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located;
The central processing unit is also used for screening a plurality of seed points and alternative points from the current frame tower point cloud;
the CPU is also used for continuously obtaining a plurality of alternative points and a plurality of seed points in a multi-frame manner, accumulating and combining the plurality of alternative points and the plurality of seed points to obtain a plurality of combined standby pointsSelecting a point and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
the central processing unit is further used for combining the seed point cloud cluster and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
Determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
the central processing unit is also used for traversing all the pole tower point clouds in the target pole tower point cloud cluster, calculating the pole tower bounding box volume and pole tower coordinates according to the coordinates of the pole tower point clouds, and extracting the highest point of the target pole tower point cloud cluster as the highest elevation of the pole tower center.
Correspondingly, the invention also provides a robust unmanned aerial vehicle laser radar line-imitating tower recognition method, which comprises the following operation steps:
acquiring all power line channel data of a current frame and all tower point clouds of the current frame in real time, and acquiring the tower point clouds of the current frame from all the tower point clouds of the current frame;
screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located;
screening a plurality of seed points and alternative points from the current frame tower point cloud;
A plurality of alternative points and a plurality of seed points are obtained through continuous multiframes, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
presetting a minimum overlapping rate threshold k, and merging the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
Calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
traversing all point clouds in the target tower point cloud cluster, calculating the volume of the tower bounding box and the coordinates of the tower, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the center of the tower.
Compared with the prior art, the embodiment of the invention has at least the following technical advantages:
according to analysis of the robust unmanned aerial vehicle laser radar line-simulating tower identification method and device provided by the invention, when the method and device are specifically applied, firstly, the power line channel and the tower of each frame of data are extracted in the flight process, all the power line channel data and all the current frame of tower point clouds of the current frame are obtained in real time according to the laser radar module, and all the tower point clouds of the current frame are obtained; the method comprises the steps of screening and determining that one power line channel in which the current frame tower point cloud is located is the power line channel in which the current frame tower point cloud is located, screening out the influence of other power line channels on tower identification through the improved operation mode, and further determining the power line channel in which the current frame tower point cloud is located; on the basis of determining a power line channel where the current frame tower point cloud is located, screening a plurality of seed points and alternative points from the current frame tower point cloud; screening out point clouds of other objects (tree point clouds and column-shaped ground point clouds similar to a pole tower), and screening out pole tower point clouds of the identified pole tower, so that the identification of the current pole tower is ensured to be accurate;
Then, a plurality of alternative points and a plurality of seed points are obtained continuously for multiple frames, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; in the prior art, the laser radar scans the same position for a plurality of times to obtain a plurality of repeated tower point clouds, so that the point clouds are gathered, and the technical scheme provided by the embodiment of the invention samples the obtained plurality of alternative points and seed points to obtain a small amount of but insufficient tower point clouds, so that the recognition accuracy of the towers is improved;
clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters; in the prior art, when a laser radar scans the point cloud of a tower, the point cloud of column-like ground objects and branches of the similar tower beside the tower is scanned together, but in general, space isolation exists between the tower and the column-like ground objects and branches of the similar tower beside the tower, and after a value is given, a DBSCAN can gather the point cloud clusters to obtain the point cloud clusters;
Combining the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box; calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster; determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster; calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster; in the embodiment of the invention, the seed point cloud cluster is a subset of the candidate point cloud clusters, so that the overlapping rate of the seed point cloud cluster bounding box and the bounding box of the candidate point cloud cluster is quite large, and a plurality of effective tower point cloud clusters can be obtained by the technical scheme provided by the embodiment of the invention;
and finally traversing all tower point clouds in the target tower point cloud cluster, calculating the volume of the tower bounding box and the tower coordinates according to the coordinates of the tower point clouds, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the tower center.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a robust unmanned aerial vehicle laser radar line-imitating tower recognition device according to an embodiment of the present invention;
fig. 2 is a schematic operation flow diagram of a robust unmanned aerial vehicle laser radar line-imitating tower recognition method according to a second embodiment of the present invention;
fig. 3 is a schematic diagram of an operation flow of a power line channel in which a current frame tower point cloud is screened and determined to be the power line channel in which the current frame tower point cloud is located in the robust unmanned aerial vehicle laser radar line-simulating tower identification method according to the second embodiment of the present invention;
fig. 4 is a schematic diagram of an operation flow of a power line channel in which a current frame tower point cloud is screened and determined to be the power line channel in which the current frame tower point cloud is located in the robust unmanned aerial vehicle laser radar line-simulating tower identification method according to the second embodiment of the present invention;
Fig. 5 is a schematic diagram of an operation flow for screening out a plurality of seed points and alternative points in a robust unmanned aerial vehicle laser radar line-simulating tower recognition method according to a second embodiment of the present invention;
fig. 6 is a schematic diagram of an operation flow for obtaining multiple candidate point cloud clusters and seed point cloud clusters in a robust unmanned aerial vehicle laser radar line-simulating tower recognition method according to a second embodiment of the present invention.
Reference numerals: a line-simulated flying unmanned aerial vehicle 10; a lidar module 11; a central processing unit 12; a GPS positioning module 13.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention will now be described in further detail with reference to specific examples thereof in connection with the accompanying drawings.
Example 1
Referring to fig. 1, the invention provides a robust unmanned aerial vehicle laser radar line-imitating tower recognition device, which comprises a line-imitating flying unmanned aerial vehicle 10;
The line-imitating flying unmanned aerial vehicle comprises a laser radar module 11, a central processing unit 12 and a GPS positioning module 13;
the central processing unit 12 is respectively connected with the laser radar module 11 and the GPS positioning module 13;
the laser radar module 11 is used for acquiring all power line channel data of a current frame and a tower point cloud of the current frame in real time;
the current frame power line channel data comprises line linear equations of all power line channels of the current frame; the pole tower point cloud comprises coordinates of the pole tower point cloud and the height of the pole tower point cloud;
the GPS positioning module 13 is used for acquiring the coordinate position of the line-simulated flying unmanned aerial vehicle in real time;
the central processing unit 12 is configured to obtain all power line channel data of a current frame and all tower point clouds of the current frame in real time according to the laser radar module, and obtain the tower point clouds of the current frame from all the tower point clouds of the current frame; screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located;
the central processing unit is also used for screening a plurality of seed points and alternative points from the current frame tower point cloud;
the central processing unit is further used for continuously obtaining a plurality of alternative points and a plurality of seed points in a multi-frame mode, accumulating and combining the plurality of alternative points and the plurality of seed points to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
the central processing unit is further used for combining the seed point cloud cluster and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
the central processing unit is also used for traversing all the pole tower point clouds in the target pole tower point cloud cluster, calculating the pole tower bounding box volume and pole tower coordinates according to the coordinates of the pole tower point clouds, and extracting the highest point of the target pole tower point cloud cluster as the highest elevation of the pole tower center.
In summary, the robust unmanned aerial vehicle laser radar line-simulating tower recognition device provided by the invention acquires all power line channel data of a current frame and a current frame tower point cloud through the laser radar module; the coordinate position of the line-simulated flying unmanned aerial vehicle is obtained in real time through the GPS positioning module; acquiring the tower point clouds of the current frame from all the tower point clouds of the current frame; screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located; further screening a plurality of seed points and alternative points from the current frame tower point cloud;
a plurality of alternative points and a plurality of seed points are obtained through continuous multiframes, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
Combining the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
and finally traversing all tower point clouds in the target tower point cloud cluster, calculating the volume of the tower bounding box and the tower coordinates according to the coordinates of the tower point clouds, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the tower center.
Example two
As shown in fig. 2, correspondingly, the invention also provides a robust unmanned aerial vehicle laser radar line-simulating tower recognition method, which comprises the following operation steps:
Step S10: acquiring all power line channel data of a current frame and all tower point clouds of the current frame in real time, and acquiring the tower point clouds of the current frame from all the tower point clouds of the current frame; (the tower point cloud refers to a tower point cloud on a current frame of power line channels, and the current frame generally contains a plurality of power line channels, countless tower point clouds and other object point clouds, wherein the tower point clouds comprise coordinates of the tower point clouds and heights of the tower point clouds);
when the unmanned aerial vehicle is initialized to perform laser radar line-imitating flight, acquiring power line channel data and tower point clouds in real time through the unmanned aerial vehicle laser radar, extracting a power line channel and point clouds (comprising the tower point clouds and the like) of each frame, respectively analyzing the power line channel data and the tower point clouds of each frame, and detailing subsequent operation;
wherein the current frame of power line channel data includes line linear equations (typically, each power line channel has a plurality of lines, and each line can be represented by a line linear equation) of all power line channels of the current frame;
the pole tower point clouds comprise pole tower point clouds and columnar ground point clouds on the current frame power line channel (the columnar ground point clouds refer to ground point clouds similar to pole towers of the power line channel, and meanwhile, the columnar ground point clouds are also referred to as rod-like body point clouds in the industry);
Step S20: screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located;
specifically, referring to fig. 3, in step S20, the screening determines that one power line channel where the current frame tower point cloud is located is the power line channel where the current frame tower point cloud is located, including the following operation steps:
step S21, obtaining line straight line equations in all power line channel data of a current frame;
step S22: storing the heights of the line linear equations into a container and sequencing to obtain a power line height sequence table; obtaining the minimum height H of the power line channel line through the power line height sequence table min And maximum height H max ；
Step S23: preset minimum vertical spacing threshold delta min Threshold delta from maximum vertical spacing max Calculating a line vertical interval average value according to the power line height sequence table; according to the average value of the vertical distance of the circuit and the minimum vertical distance threshold delta min Threshold delta from maximum vertical spacing max Calculating and determining a 2D distance threshold D between each power line channel 2d And a 3D distance threshold D between each power line channel 3d ；
Acquiring a current frame tower point cloud from all tower point clouds of the current frame, calculating 2D distances and 3D distances between the current frame tower point cloud and all power line channels of the current frame respectively, screening that the 2D distances between the current frame tower point cloud and all power line channels of the current frame respectively are smaller than a 2D distance threshold D between each power line channel 2d And the pole tower point clouds of the current frame are respectively connected to all power line channels of the current frameThe 3D distance is less than a 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; and determining one power line channel where the screened current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located.
It should be noted that, there is a line equation for each power line channel, and the line equation may reflect the height of each frame of each power line channel.
Specifically, referring to fig. 4, in step S23, the preset minimum vertical spacing threshold δ min Threshold delta from maximum vertical spacing max Calculating a line vertical interval average value according to the power line height sequence table; according to the average value of the vertical distance of the circuit and the minimum vertical distance threshold delta min Threshold delta from maximum vertical spacing max Calculating and determining a 2D distance threshold D between each power line channel 2d And a 3D distance threshold D between each power line channel 3d The method comprises the steps of carrying out a first treatment on the surface of the Calculating 2D distances and 3D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively, and screening that the 2D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively are smaller than a 2D distance threshold D between each two power line channels 2d And the 3D distances from the tower point cloud of the current frame to all the power line channels of the current frame are respectively smaller than the 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; determining that one power line channel where the screened current frame tower point cloud is located is the power line channel where the current frame tower point cloud is located, including the following operation steps:
step S231: preset minimum vertical spacing threshold delta min The method comprises the steps of carrying out a first treatment on the surface of the Traversing the heights of all the line linear equations in the power line height sequence table, calculating the absolute height difference of the heights of two adjacent line linear equations in the container, and if the absolute height difference is larger than the minimum vertical distance threshold delta min Determining the absolute height difference as an effective height difference; calculating a plurality of the effective height differences (namely, a plurality of (absolute height differences of the heights of two adjacent line straight line equations))Obtaining a line vertical interval average value by the average value;
step S232: preset maximum vertical spacing threshold delta max The method comprises the steps of carrying out a first treatment on the surface of the Judging the average value of the vertical distance of the circuit and the threshold delta of the maximum vertical distance max If the line vertical pitch average is greater than the maximum vertical pitch threshold delta max Determining that the 2D distance threshold value between each power line channel and the 3D distance threshold value between each power line channel are delta max (i.e. D 2d ＝δ max ，D 3d ＝δ max ) The method comprises the steps of carrying out a first treatment on the surface of the If the current line vertical spacing average value is smaller than the maximum vertical spacing threshold delta max The 2D distance threshold value between each power line channel and the 3D distance threshold value between each power line channel are delta min (i.e. D 2d ＝δ min ，D 3d ＝δ min )；
Calculating 2D distances and 3D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively, and screening that the 2D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively are smaller than a 2D distance threshold D between each two power line channels 2d And the 3D distances from the tower point cloud of the current frame to all the power line channels of the current frame are respectively smaller than the 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; and determining one power line channel where the screened current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located.
The 2D distance threshold D between each power line channel 2d And a 3D distance threshold D between each power line channel 3d Is adjusted according to the change of the spacing between the lines.
Illustrating: one point cloud of the tower of the current frame is arranged, and three power line channels of the current frame are respectively represented as a power line channel 1, a power line channel 2 and a power line channel 3; the heights of the line linear equations of the three power line channels are 1 meter, 5 meters and 11 meters respectively, and are ordered in a power line height sequence table according to the heights; the preset minimum vertical spacing threshold value is 1 meter, the line linear equation heights of three power line channels are calculated according to the heights of two adjacent line linear equations in the power line height sequence table, namely, the absolute height differences are respectively 4 meters and 6 meters, and at the moment, the absolute height difference is larger than the minimum vertical spacing threshold value by 1 meter, and then the two absolute height differences are effective height differences;
Calculating an average value of the two effective height differences to obtain a line vertical distance average value of 5 meters ((4 meters+6 meters)/2=5 meters);
a maximum vertical spacing threshold value is preset for 3 meters, at the moment, the line vertical spacing average value is larger than the maximum vertical spacing threshold value for 3 meters, 2D and 3D distance thresholds from the tower point cloud to three power line channels are determined to be the maximum vertical spacing threshold value for 3 meters, and one power line channel of the three power line channels with 2D and 3D distances from the current frame tower point cloud to the three power line channels being smaller than 3 meters is determined to be the power line channel where the current frame tower point cloud is located;
if the average value of the vertical distance of the line is 2 meters and is smaller than the maximum vertical distance threshold value by 3 meters, determining that the 2D and 3D distance thresholds of the tower point cloud to the three power line channels are both 2 meters of the average value of the vertical distance of the line, and determining that one power line channel with the 2D and 3D distances of the current frame tower point cloud to the three power line channels being smaller than 2 meters is the power line channel where the current frame tower point cloud is located.
Because different lines have different sizes, the line spacing and the line-to-line height are greatly different, when the line linear equation (the line linear equation of the current frame is utilized) is utilized for filtering, the distance threshold parameter needs to be dynamically adjusted according to the actual line condition, otherwise, incomplete filtering of the pole tower is easy to cause; for a distribution network (or called a distribution network line), as the towers of the distribution network line are relatively small and the distance between adjacent lines in the distribution network is small, two values meeting actual line specifications can be directly specified, so that the towers are still complete after filtration, and additional calculation is not needed; for a main network, the range of a tower is large, and self-adaptive estimation is required to be carried out by utilizing the distance of a line; the set point to line 2D and 3D distances are D, respectively 2d 、D 3d ；
The point is a coordinate point (or coordinate point of point cloud) or a power line channel line in the space; the 2D distance from a point to a line is the perpendicular distance from the point to the line projected onto the two-dimensional plane XY. The 3D distance from the point to the line, that is, the vertical distance from the space three-dimensional point to the space straight line, and the "line" is a straight line segment obtained by fitting the power line point cloud, and can be represented by a space straight line equation.
If the absolute height difference is smaller than the given minimum pitch threshold δ min The level difference is not taken as an effective value, and a split wire or a line with the other side being close in height can exist in the line; therefore, direct filtering does not take part in the computation.
If the absolute height difference is greater than a given minimum spacing threshold delta min The current height difference is considered to be valid and used for average calculation; all effective height differences are averaged to obtain a line vertical pitch average (i.e., an estimated line vertical pitch average).
It should be noted that some lines may be estimated to have deviation, such as the line itself may be incomplete or the algorithm may have line loss (very low probability) when extracting the line, and we can know a maximum range according to some line designs, if the estimation exceeds the range, it is noted that the estimation has problems, and the maximum value is expressed and is enough.
Step S30: screening a plurality of seed points and alternative points from the current frame tower point cloud;
specifically, in step S30, a plurality of seed points and candidate points are screened from the current frame tower point cloud, including the following operation steps:
step S31, judging whether the tower is identified before the current frame tower point cloud is acquired, and according to a preset height threshold delta h Minimum height H of power line channel line min And maximum height H max The method comprises the steps of determining an included angle between the direction from a coordinate point of a current line-imitating flying unmanned aerial vehicle to a point cloud of a current pole and the flying direction of the current line-imitating flying unmanned aerial vehicle, 2D distance and 3D distance between the point cloud of the pole and a power line, a 2D distance threshold value and a 3D distance threshold value between lines, and a seed point distance threshold value D s Screening a plurality of seed points and alternative points;
specifically, referring to fig. 5, in step S31Judging whether a tower is identified before the current frame tower point cloud is acquired, and according to a preset height threshold delta h Minimum height H of power line channel line min And maximum height H max The method comprises the steps of determining an included angle between the direction from a coordinate point of a current line-imitating flying unmanned aerial vehicle to a point cloud of a current pole and the flying direction of the current line-imitating flying unmanned aerial vehicle, 2D distance and 3D distance between the point cloud of the pole and a power line, a 2D distance threshold value and a 3D distance threshold value between lines, and a seed point distance threshold value D s Screening a plurality of seed points and alternative points, comprising the following operation steps:
step S311: randomly selecting any one point in the pole tower point cloud on the current frame power line channel as a primary target point cloud;
step S312: judging whether an identified tower exists before the tower point cloud on the current frame power line channel is acquired, if the identified tower exists before the tower point cloud on the current frame power line channel is acquired, acquiring bounding box information of the last tower, judging whether primary target point cloud is in the bounding box, and if the primary target point cloud is in the bounding box, filtering the primary target point cloud and reselecting the primary target point cloud; if the primary target point cloud is not in the current bounding box, determining that the current primary target point cloud is a secondary target point cloud; if no identified tower exists before the tower point cloud on the current frame of power line is acquired, determining that the primary target point cloud is a secondary target point cloud;
illustrating: a, B, C, D tower point clouds exist in the previous tower bounding box, the current primary target point cloud is E, the current primary target point cloud E is determined to be the secondary target point cloud, and the next operation is carried out; if the current primary target point cloud is D, screening and filtering the current primary target point cloud, and performing no further operation on the current primary target point cloud;
When the unmanned aerial vehicle flies like a line, the laser radar scans the flown towers, the flown towers are recorded in the historical tower information, and the flown towers need to be filtered.
Step S313: preset height threshold delta h The method comprises the steps of carrying out a first treatment on the surface of the If the secondary targetThe height of the point cloud is greater than H max +Δ h Or less than H min -Δ h Filtering the cloud and reselecting a secondary target point cloud; if the height of the secondary target point cloud is smaller than H max +Δ h And is greater than H min -Δ h Determining the secondary target point cloud as a tertiary target point cloud;
illustrating: the preset height threshold is 1 meter, the minimum height of the line linear equation of the power line channel is 3 meters, and the maximum height of the line linear equation of the power line channel is 10 meters, so that the height of the secondary target point cloud (tower point cloud) needs to be H min -Δ h Above (i.e. the height of the secondary target point cloud is greater than H min -Δ h ) Screening out the point cloud of trees below the power line channel or column-like ground objects similar to a pole tower; the height of the secondary target point cloud needs to be H max +Δ h The following (i.e. the height of the secondary target point cloud is less than H max +Δ h ) Screening out the pole tower point clouds on other power line channels above the current point route channel; i.e. if the height of the secondary target point cloud is smaller than H max +Δ h And the height of the secondary target point cloud is greater than H min -Δ h Determining the secondary target point cloud as a tertiary target point cloud and performing the next operation;
h is the same as min And H max Is the lowest and highest value of the line channel height, where it is understood that the algorithm only requires points within a certain range above and below the line channel vertical height, and that a large probability of exceeding the range is other disturbances, such as other lines above, or trees below, etc.
Step S314: acquiring the direction from the position coordinate of the current unmanned aerial vehicle to the three-level target point cloud coordinate, judging whether an included angle between the direction and the current unmanned aerial vehicle line-simulating flight direction (the included angle between the direction and the current unmanned aerial vehicle line-simulating flight direction is a formed by taking the ground as a plane, and mapping the direction and the current unmanned aerial vehicle line-simulating flight direction on the plane) exceeds 90 degrees, and determining the three-level target point cloud as a four-level target point cloud if the included angle between the direction and the current unmanned aerial vehicle line-simulating flight direction does not exceed 90 degrees;
illustrating: taking the position coordinate of the current unmanned aerial vehicle as a starting point a, taking the three-level target point cloud as an end point b, and calculating the direction to be expressed as Calculating the degree of an included angle of less than ab between the direction and the current flight direction of the unmanned aerial vehicle, and screening out the three-level target point cloud and not processing the three-level target point cloud if the degree of the included angle of less than ab is greater than 90 degrees and the three-level target point cloud is behind the flight direction of the unmanned aerial vehicle; if the degree of the included angle < ab > is smaller than 90 degrees, the three-level target point cloud is in front of the current unmanned aerial vehicle flight direction, the three-level target point cloud is determined to be a four-level target point cloud, and the next operation is carried out;
the direction filtering was performed, and only the point in front of the flight direction was valid; when the unmanned aerial vehicle flies in a line-imitating mode, the laser radar scans a flown tower, the flown tower is recorded in historical tower information, and the flown tower needs to be filtered; calculating a direction by using the current position coordinate of the unmanned aerial vehicle and the point coordinate, determining whether the direction from the unmanned aerial vehicle to the tower is consistent with the flight direction, if so, immediately judging whether the direct included angle between the current position coordinate and the point coordinate exceeds 90 degrees, if so, indicating that the tower point is behind the unmanned aerial vehicle, otherwise, in front; the center of the tower can be extracted as a coordinate point, the current position of the unmanned aerial vehicle is a coordinate point, two coordinates can form a vector, and in addition, the flying direction is known and is also a vector, so that the size of the included angle can be calculated according to an included angle cosine formula of the two vectors;
Here we generally calculate only a two-dimensional plane, regardless of vertical direction, and after all only determine whether the aircraft is currently in front of or behind the tower. We will complete the tower identification before the tower, and if the aircraft has flown through the tower, then there is no need at this time to scan the point cloud of the tower even though the laser is still scanning, which tower is already recorded in the historical tower information, and this point cloud interference should be eliminated.
Step S315: determining a 2D distance D from the tower point cloud to the power line 2d 2D distance threshold for alternative points, 3D distance D of tower point cloud to power line 3d 3D distance threshold value for the candidate point is preset, and seed point distance threshold value D is preset s The method comprises the steps of carrying out a first treatment on the surface of the Calculating the 2D distance and the 3D distance between the four-level target point cloud and the power line, if the 2D distance between the four-level target point cloud and the power line is smaller than or equal to the 2D distance threshold D of the candidate point 2d And the 3D distance from the four-level target point cloud to the power line is smaller than or equal to the candidate point 3D distance threshold D 3d Determining the cloud of the four-level target points as the candidate points; if the 2D and 3D distances from the four-level target point cloud to the power line are simultaneously smaller than or equal to the seed point distance threshold D s Determining the fourth-level target point cloud as a seed point;
illustrating: presetting a seed point distance threshold value to be 0.5 meter, wherein a 2D distance threshold value of the alternative point is 3 meters, and a 3D distance threshold value of the alternative point is 5 meters; 5 four-level target point clouds are respectively expressed as a point Q, a point W, a point E, a point R and a point T; respectively calculating 2D and 3D distances from a point Q, a point W, a point E, a point R and a point T to a current power line channel, wherein the distances are respectively expressed as a point Q (0.1 meter, 0.3 meter), a point W (0.4 meter, 0.5 meter), a point E (6 meters, 7 meters), a point R (1 meter, 4 meters) and a point T (3 meters, 6 meters), and the point Q is determined to be a seed point because the distance between the point Q and the point W is smaller than a seed point distance threshold value by 0.5 meter and the distance between the point T and the point T is smaller than a seed point distance threshold value by 0.5 meter; if 0.4 m is less than the seed point distance threshold by 0.5 m and 0.5 m is equal to the seed point distance threshold by 0.5 m, determining the point W as a seed point; if 6 meters are larger than the 2D distance threshold value of the alternative point by 3 meters and 7 meters are larger than the 3D distance threshold value of the alternative point, screening out the point E and not processing any more; if 1 meter is smaller than the 2D distance threshold value of the alternative point by 3 meters and 4 meters is smaller than the 3D distance threshold value of the alternative point by 5 meters, determining the point R as the alternative point; 3 meters and the like and the alternative point 2D distance threshold value, but 6 meters are larger than the alternative point 3D threshold value, the point T is screened out and is not processed any more; the above Q (0.1 m, 0.3 m) means that the 2D distance from the four-level target point cloud Q to the current power line channel is 0.1 m, and the 3D distance from the four-level target point cloud Q to the current power line channel is 0.3 m; the representations of W (0.4 m, 0.5 m), E (6 m, 7 m), R (1 m, 4 m) and T (3 m, 6 m) are consistent with the representations of Q (0.1 m, 0.3 m), and the descriptions are omitted.
The dot isFiltering the shortest distance to the line; and setting different filtering thresholds to divide the tower point cloud into a tower point, an alternative point and an invalid point. Estimating the distance D in the method 2d And D 3d As alternative point threshold values, a smaller distance threshold value D is additionally set s As a seed point threshold, only points smaller than the threshold may be used as seed points. Traversing all line linear equations in the channel, calculating 2D and 3D distances from points to all lines, and taking the minimum distance for judgment; minimum distance is at the same time D 2d And D 3d Points within as alternative points, at D s The points in the tree are used as seed points;
the seed points are points which are more in line with the towers relative to the alternative points, if the distances from the line on the channel are very close, such as less than 0.5 meter, the points are basically on the line, the actual line towers are necessarily connected with the line, and other interference points such as tower points adjacent to the line, tree, light pole and the like are almost unlikely to be interfered; the alternative point threshold is set to be larger, for example, 5 meters, so that point clouds, trees or other interference on the peripheral lines can possibly be added; if seed points exist, the fact that a tower exists is meant to be truly, more alternative points are that in order to restore the shape of the tower more comprehensively, a smaller threshold value can cause the tower to be incomplete, in addition, the tower size in a distribution network line is changed in a large range, the radius is only 1 meter, and the radius is possibly 4 meters, and in the actual situation, a large range is required to be filtered; if both the seed point and the alternative point are overlapped, the actual shape of the tower is restored by utilizing the overlapping relation of the seed point and the alternative point by utilizing whether the seed point exists in the tower or not.
Step S40: a plurality of alternative points and a plurality of seed points are obtained through continuous multiframes, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting the seed pointsYun Cu minimum threshold N for point cloud number s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
specifically, referring to fig. 6, in step S40, a plurality of candidate points and a plurality of seed points are obtained in a continuous multi-frame manner, and the plurality of candidate points and the plurality of seed points are accumulated and combined to obtain a plurality of combined candidate points and seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Obtaining a plurality of candidate point cloud clusters and seed point cloud clusters, comprising the following operation steps:
Step S41: adding a plurality of alternative points and a plurality of seed points which are obtained by continuous multiframes into a history point set to accumulate and combine, so as to obtain a plurality of combined alternative points and a plurality of combined seed points; the history point set is a set formed by accumulating alternative points acquired by each frame and seed points, namely, the accumulated combination is put into the history point set, and the combination is the meaning of the placement and has no other operation meaning;
step S42: respectively collecting different multiple alternative points and multiple seed points for the multiple combined alternative points and the multiple combined seed points by utilizing voxel grid filtration (or establishing kdtree filtration), and screening out repeated alternative points and seed points to obtain sampled seed points and alternative points;
it should be noted that, the sampling process may be performed on the seed points after merging and the candidate points after merging, and the processing may be performed by using a common method such as voxel grid or kdtree establishment (where the voxel grid or kdtree establishment is in the prior art and is not described in detail), so as to obtain the seed points after sampling and the candidate points.
The voxel grid or the kdtree is established, so that judgment errors caused by the condition that some points are too dense can be prevented; because we determine whether the towers exist through the number of the seed points, the number of the towers still reaches after sampling, which indicates that the actual volume of the point cloud is larger and the volume of the towers is larger, the method is reasonable; conversely, if not sampled, the laser may scan the same location multiple times, resulting in point cloud aggregation, a very small range aggregating a large number of repeated points, and sometimes a small branch may be identified as a tower next to the line; after sampling, the number of points also represents the real occupied space of the target, so that the interference can be reduced.
Step S43: clustering the sampled candidate points by using a DBSCAN density clustering method to obtain a plurality of candidate point cloud clusters; clustering the sampled seed points by using a DBSCAN density clustering method to obtain a plurality of seed point cloud clusters to be determined;
it should be noted that, the DBSCAN density clustering method is in the prior art, and is not described in detail;
the DBSCAN is a conventional density clustering method, and because the range of the alternative points is large, the tower points, tree points and tower points of other lines of the current channel can exist in the DBSCAN; because the clearance distance exists in the current line, namely the tower of the current line is inevitably greatly separated from the point cloud of the surrounding environment or the point cloud of the adjacent line in space; after a smaller proper space distance threshold is given, the DBSCAN can automatically cluster the peripheral point cloud clusters to obtain a plurality of point cloud clusters; the seed points are small in parameter setting, the distance between the priority line and the line is wide, so that a plurality of seed clusters exist, the final result is not affected, and only the point with the largest number is selected; seed points are a subset of alternative points; if a certain cluster in the seed points and the alternative points is overlapped, the cluster is the required tower point cloud cluster, and other interference clusters are filtered by certain trees, street lamps or tower point cloud clusters adjacent to the line.
Step S44: presetting a minimum threshold N of point cloud numbers in a seed point cloud cluster s Judging whether the number of point clouds in the seed point cloud cluster to be determined is larger than a minimum threshold value N of the number of point clouds in the seed point cloud cluster s If yes, determining the seed point cloud cluster to be determined as a target seed point cloud cluster to be selected; screening target seed point cloud clusters to be selected with the largest number of point clouds as seed point cloud clusters;
it should be noted that, if at least one seed point cloud cluster is not collected after the seed point clustering, or the number of point clouds in the largest point cloud cluster is smaller than the threshold value N s Then the current frame is considered to still not have a tower, the current seed point cloud cluster can be determined to be a tower-like ground object point cloud, and the current seed point cloud cluster is screened out and is not processed any more; otherwise, taking the point cloud cluster with the largest number of points as a final seed point cloud cluster; the seed point can be used as a very close position of a certain line, because each channel is provided with a plurality of lines, the set distance is too small, the line spacing is relatively large, for example, the set distance from the seed point to the line is 1 meter, the actual line spacing is 5 meters, and thus each line can have a seed point set; whether the target seed point cloud cluster is selected, the target seed point cloud cluster with the largest number of the selected point clouds can be selected as the seed point cloud cluster by preferentially selecting the target seed point cloud cluster with the largest overlapping (namely, the target seed point cloud cluster with the largest number of the selected point clouds) which is more stable, as long as the overlapping of the target seed point cloud cluster and the current pole tower reaches a threshold value; and especially near the boundary, the calculated overlap ratio is relatively smaller than that of other towers.
Step S50: presetting a minimum overlapping rate threshold k, and merging the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes (bounding box coverage) of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the seed point cloud cluster bounding box is overlapped with the candidate point cloud cluster bounding box, and if so, calculating the overlapping rate of the seed point cloud cluster bounding box and the candidate point cloud cluster bounding box;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
if the candidate points are overlapped (the seed point cloud cluster bounding box is overlapped with the candidate point cloud cluster bounding box), the candidate points and the seed points are overlapped, and the current candidate point cloud cluster is used for restoring the shape of the pole tower and is used as an effective pole tower point cloud cluster; setting an overlapping rate K, if the plane overlapping rate of the candidate point cloud cluster and the seed point cloud cluster is smaller than K, invalidating the candidate point cloud cluster, otherwise, adding the candidate point cloud cluster into the effective pole tower point cloud cluster.
Step S60: calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster (the superposition plane area refers to the superposition area of the candidate point cloud cluster corresponding to the effective tower point cloud cluster and the seed point cloud cluster mapped to a plane formed by the two points together), and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
traversing all point clouds in the target tower point cloud cluster, calculating the volume (volume) of the tower bounding box and the tower coordinates, extracting the highest point of the target tower point cloud cluster as the highest elevation of the tower center ("calculating the volume (volume) of the tower bounding box and the tower coordinates, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the tower center" as determining the volume, the position and the highest point of the tower to be determined).
In summary, the method and the device for identifying the simulated line towers of the unmanned aerial vehicle laser radar provided by the embodiment of the invention can be known that, when the method and the device are specifically applied, firstly, the power line channel and the towers of each frame of data are extracted in the flight process, all the power line channel data and all the current frame of tower point clouds of the current frame are obtained in real time according to the laser radar module, and all the tower point clouds of the current frame are obtained; the method comprises the steps of screening and determining that one power line channel in which the current frame tower point cloud is located is the power line channel in which the current frame tower point cloud is located, screening out the influence of other power line channels on tower identification through the improved operation mode, and further determining the power line channel in which the current frame tower point cloud is located; on the basis of determining a power line channel where the current frame tower point cloud is located, screening a plurality of seed points and alternative points from the current frame tower point cloud; screening out point clouds of other objects (tree point clouds and column-shaped ground point clouds similar to a pole tower), and screening out pole tower point clouds of the identified pole tower, so that the identification of the current pole tower is ensured to be accurate;
Then, a plurality of alternative points and a plurality of seed points are obtained continuously for multiple frames, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; in the prior art, the laser radar scans the same position for a plurality of times to obtain a plurality of repeated tower point clouds, so that the point clouds are gathered, and the technical scheme provided by the embodiment of the invention samples the obtained plurality of alternative points and seed points to obtain a small amount of but insufficient tower point clouds, so that the recognition accuracy of the towers is improved;
clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters; in the prior art, when a laser radar scans the point cloud of a tower, the point cloud of column-like ground objects and branches of the similar tower beside the tower is scanned together, but in general, space isolation exists between the tower and the column-like ground objects and branches of the similar tower beside the tower, and after a value is given, a DBSCAN can gather the point cloud clusters to obtain the point cloud clusters;
Combining the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box; calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster; determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster; calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster; in the embodiment of the invention, the seed point cloud cluster is a subset of the candidate point cloud clusters, so that the overlapping rate of the seed point cloud cluster bounding box and the bounding box of the candidate point cloud cluster is quite large, and a plurality of effective tower point cloud clusters can be obtained by the technical scheme provided by the embodiment of the invention;
and finally traversing all tower point clouds in the target tower point cloud cluster, calculating the volume of the tower bounding box and the tower coordinates according to the coordinates of the tower point clouds, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the tower center.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; modifications of the technical solutions described in the foregoing embodiments, or equivalent substitutions of some or all of the technical features thereof, may be made by those of ordinary skill in the art; such modifications and substitutions do not depart from the spirit of the invention.

Claims (10)

1. A robust unmanned aerial vehicle laser radar line-simulating pole tower identification method is characterized by comprising the following operation steps:
acquiring all power line channel data of a current frame and all tower point clouds of the current frame in real time, and acquiring the tower point clouds of the current frame from all the tower point clouds of the current frame;
screening and determining one power line channel where the current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located;
screening a plurality of seed points and alternative points from the current frame tower point cloud;
a plurality of alternative points and a plurality of seed points are obtained through continuous multiframes, and the plurality of alternative points and the plurality of seed points are accumulated and combined to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
presetting a minimum overlapping rate threshold k, and merging the seed point cloud cluster clustering result and the alternative point cloud cluster clustering result to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
traversing all point clouds in the target tower point cloud cluster, calculating the volume of the tower bounding box and the coordinates of the tower, and extracting the highest point of the target tower point cloud cluster as the highest elevation of the center of the tower.
2. The method for identifying the simulated line tower of the unmanned aerial vehicle laser radar according to claim 1, wherein the step of screening determines that one power line channel where the point cloud of the tower of the current frame is the power line channel where the point cloud of the tower of the current frame is located, specifically comprises the following steps:
Acquiring line linear equations in all power line channel data of a current frame;
storing the heights of the line linear equations into a container and sequencing to obtain a power line height sequence table; obtaining the minimum height H of the power line channel line through the power line height sequence table min And maximum height H max ；
Preset minimum vertical spacing threshold delta min Threshold delta from maximum vertical spacing max Calculating a line vertical interval average value according to the power line height sequence table; according to the average value of the vertical distance of the circuit and the minimum vertical distance threshold delta min Threshold delta from maximum vertical spacing max Computing a 2D distance threshold between each of the power line channelsValue D 2d And a 3D distance threshold D between each power line channel 3d ；
Acquiring a current frame tower point cloud from all tower point clouds of the current frame, calculating 2D distances and 3D distances between the current frame tower point cloud and all power line channels of the current frame respectively, screening that the 2D distances between the current frame tower point cloud and all power line channels of the current frame respectively are smaller than a 2D distance threshold D between each power line channel 2d And the 3D distances from the tower point cloud of the current frame to all the power line channels of the current frame are respectively smaller than the 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; and determining one power line channel where the screened current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located.
3. The method for identifying the line-imitating tower of the unmanned aerial vehicle laser radar according to claim 2, wherein the step of screening a plurality of seed points and candidate points from the current frame tower point cloud comprises the following steps:
judging whether a tower is identified before the current frame tower point cloud is acquired, and according to a preset height threshold delta h Minimum height H of power line channel line min And maximum height H max The method comprises the steps of determining an included angle between the direction from a coordinate point of a current line-imitating flying unmanned aerial vehicle to a point cloud of a current pole and the flying direction of the current line-imitating flying unmanned aerial vehicle, 2D distance and 3D distance between the point cloud of the pole and a power line, a 2D distance threshold value and a 3D distance threshold value between lines, and a seed point distance threshold value D s A plurality of seed points and alternative points are screened out.
4. A method for identifying a robust unmanned aerial vehicle lidar line-like tower according to claim 3, wherein the predetermined minimum vertical separation threshold δ is min Threshold delta from maximum vertical spacing max Calculating a line vertical interval average value according to the power line height sequence table; according to the average value of the vertical distance of the circuit and the minimum vertical distance threshold delta min Threshold delta from maximum vertical spacing max Calculating and determining a 2D distance threshold D between each power line channel 2d And a 3D distance threshold D between each power line channel 3d The method comprises the steps of carrying out a first treatment on the surface of the Calculating 2D distances and 3D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively, and screening that the 2D distances from the current frame pole and tower point cloud to all power line channels of the current frame respectively are smaller than a 2D distance threshold D between each two power line channels 2d And the 3D distances from the tower point cloud of the current frame to all the power line channels of the current frame are respectively smaller than the 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; determining that one power line channel where the screened current frame tower point cloud is located is the power line channel where the current frame tower point cloud is located, including the following operation steps:
preset minimum vertical spacing threshold delta min The method comprises the steps of carrying out a first treatment on the surface of the Traversing the heights of all the line linear equations in the power line height sequence table, calculating the absolute height difference of the heights of two adjacent line linear equations in the container, and if the absolute height difference is larger than the minimum vertical distance threshold delta min Determining the absolute height difference as an effective height difference; calculating average values of the effective height differences to obtain line vertical interval average values;
preset maximum vertical spacing threshold delta max The method comprises the steps of carrying out a first treatment on the surface of the Judging the average value of the vertical distance of the circuit and the threshold delta of the maximum vertical distance max If the line vertical pitch average is greater than the maximum vertical pitch threshold delta max Determining that the 2D distance threshold value between each power line channel and the 3D distance threshold value between each power line channel are delta max The method comprises the steps of carrying out a first treatment on the surface of the If the current line vertical spacing average value is smaller than the maximum vertical spacing threshold delta max The 2D distance threshold value between each power line channel and the 3D distance threshold value between each power line channel are delta min ；
Calculating 2D distances and 3D distances from the current frame pole tower point cloud to all power line channels of the current frame respectively, and screening that the 2D distances from the current frame pole tower point cloud to all power line channels of the current frame respectively are small2D distance threshold D between each power line channel 2d And the 3D distances from the tower point cloud of the current frame to all the power line channels of the current frame are respectively smaller than the 3D distance threshold D between each power line channel 3d The current frame tower point cloud is the screened current frame tower point cloud; and determining one power line channel where the screened current frame tower point cloud is located as the power line channel where the current frame tower point cloud is located.
5. The method for identifying a robust unmanned aerial vehicle laser radar line-simulating tower according to claim 4, wherein the determining whether a tower has been identified before the current frame of tower point cloud is obtained is based on a preset height threshold delta h Minimum height H of power line channel line min And maximum height H max The method comprises the steps of determining an included angle between the direction from a coordinate point of a current line-imitating flying unmanned aerial vehicle to a point cloud of a current pole and the flying direction of the current line-imitating flying unmanned aerial vehicle, 2D distance and 3D distance between the point cloud of the pole and a power line, a 2D distance threshold value and a 3D distance threshold value between lines, and a seed point distance threshold value D s Screening a plurality of seed points and alternative points, comprising the following operation steps:
randomly selecting any one point in the pole tower point cloud on the current frame power line channel as a primary target point cloud;
judging whether an identified tower exists before the tower point cloud on the current frame power line channel is acquired, if the identified tower exists before the tower point cloud on the current frame power line channel is acquired, acquiring bounding box information of the last tower, judging whether primary target point cloud is in the bounding box, and if the primary target point cloud is in the bounding box, filtering the primary target point cloud and reselecting the primary target point cloud; if the primary target point cloud is not in the current bounding box, determining that the current primary target point cloud is a secondary target point cloud; if no identified tower exists before the tower point cloud on the current frame of power line is acquired, determining that the primary target point cloud is a secondary target point cloud;
Preset height threshold delta h The method comprises the steps of carrying out a first treatment on the surface of the If the height of the secondary target point cloud is greater than H max +Δ h Or less than H min -Δ h Filtering the cloud and reselecting a secondary target point cloud; if the height of the secondary target point cloud is smaller than H max +Δ h And is greater than H min -Δ h Determining the secondary target point cloud as a tertiary target point cloud;
acquiring the direction from the position coordinate of the current unmanned aerial vehicle to the three-level target point cloud coordinate, judging whether an included angle between the direction and the current unmanned aerial vehicle line-simulating flight direction exceeds 90 degrees, and if the included angle between the direction and the current unmanned aerial vehicle line-simulating flight direction does not exceed 90 degrees, determining the three-level target point cloud as a four-level target point cloud;
determining a 2D distance D from the tower point cloud to the power line 2d 2D distance threshold for alternative points, 3D distance D of tower point cloud to power line 3d 3D distance threshold value for the candidate point is preset, and seed point distance threshold value D is preset s The method comprises the steps of carrying out a first treatment on the surface of the Calculating the 2D distance and the 3D distance between the four-level target point cloud and the power line, if the 2D distance between the four-level target point cloud and the power line is smaller than or equal to the 2D distance threshold D of the candidate point 2d And the 3D distance from the four-level target point cloud to the power line is smaller than or equal to the candidate point 3D distance threshold D 3d Determining the cloud of the four-level target points as the candidate points; if the 2D and 3D distances from the four-level target point cloud to the power line are simultaneously smaller than or equal to the seed point distance threshold D s And determining the fourth-level target point cloud as a seed point.
6. The method for identifying the line-imitating tower of the unmanned aerial vehicle laser radar, which is characterized by comprising the steps of obtaining a plurality of candidate points and a plurality of seed points by continuous multiframes, accumulating and combining the candidate points and the seed points to obtain a plurality of combined candidate points and seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters, and packagingThe method comprises the following operation steps:
adding a plurality of alternative points and a plurality of seed points which are obtained by continuous multiframes into a history point set to accumulate and combine, so as to obtain a plurality of combined alternative points and a plurality of combined seed points;
collecting different alternative points and seed points for the combined alternative points and the combined seed points by utilizing voxel grid filtration, and screening out repeated alternative points and seed points to obtain sampled seed points and alternative points;
Clustering the sampled candidate points by using a DBSCAN density clustering method to obtain a plurality of candidate point cloud clusters; clustering the sampled seed points by using a DBSCAN density clustering method to obtain a plurality of seed point cloud clusters to be determined;
presetting a minimum threshold N of point cloud numbers in a seed point cloud cluster s Judging whether the number of point clouds in the seed point cloud cluster to be determined is larger than a minimum threshold value N of the number of point clouds in the seed point cloud cluster s If yes, determining the seed point cloud cluster to be determined as a target seed point cloud cluster to be selected; and screening the target seed point cloud clusters to be selected with the largest number of point clouds as seed point cloud clusters.
7. The method for identifying the robust unmanned aerial vehicle laser radar line-imitating towers according to claim 6, wherein the bounding box information of the towers comprises all tower point clouds forming the towers.
8. The method for identifying a robust unmanned aerial vehicle lidar line-like shaft tower according to claim 6, wherein the minimum vertical separation threshold δ min Is constant; the maximum vertical spacing threshold delta max Is constant.
9. The method for identifying a robust unmanned aerial vehicle laser radar line-simulating tower according to claim 6, wherein the 2D distance threshold D between each of the power line channels 2d And a 3D distance threshold D between each power line channel 3d According to the line-to-line relationshipA distance change adjusting 2D distance threshold D between each power line channel 2d Is the minimum vertical spacing threshold delta min Or a maximum vertical spacing threshold delta max The method comprises the steps of carrying out a first treatment on the surface of the The 3D distance threshold D between each power line channel 3d Is the minimum vertical spacing threshold delta min Or a maximum vertical spacing threshold delta max The method comprises the steps of carrying out a first treatment on the surface of the The seed point distance threshold D s Is constant; the height threshold delta h Is constant; the minimum threshold value N of the point cloud number of the seed point cloud cluster s Is constant; the overlap ratio minimum threshold k is constant.
10. A robust unmanned plane laser radar line-imitating tower recognition device is characterized by comprising a line-imitating flying unmanned plane;
the line-imitating flying unmanned aerial vehicle comprises a laser radar module, a central processing unit and a GPS positioning module;
the central processing unit is respectively connected with the laser radar module and the GPS positioning module;
the laser radar module is used for acquiring all power line channel data of the current frame and the pole tower point cloud of the current frame in real time;
the current frame power line channel data comprises line linear equations of all power line channels of the current frame; the pole tower point cloud comprises coordinates of the pole tower point cloud and the height of the pole tower point cloud;
The GPS positioning module is used for acquiring the coordinate position of the line-simulated flying unmanned aerial vehicle; the central processing unit is used for obtaining a power line height sequence table through a linear equation of the current power line channel according to all power line channel data of the current frame and the current frame tower point cloud obtained in real time by the laser radar module, and further obtaining the minimum height H of the power line channel line min And maximum height H max The method comprises the steps of carrying out a first treatment on the surface of the Determining a 2D distance threshold value and a 3D distance threshold value between lines through a circuit line height sequence table; determining that one electric power first channel where the current frame tower point cloud is located is an electric power channel where the current frame tower point cloud is located;
judging that the current frame is acquiredWhether the tower is identified before the tower point cloud or not, and according to a preset height threshold delta h Minimum height H of power line channel line min And maximum height H max The included angle between the direction from the coordinate point of the current line-imitating flying unmanned aerial vehicle to the point cloud of the current pole and the flying direction of the current line-imitating flying unmanned aerial vehicle, the 2D distance and the 3D distance from the point cloud of the pole to the power line and the 2D distance threshold D between the lines 2d 3D distance threshold D 3d Seed point distance threshold D s Screening a plurality of seed points and alternative points;
The central processing unit is further used for continuously obtaining a plurality of alternative points and a plurality of seed points in a multi-frame mode, accumulating and combining the plurality of alternative points and the plurality of seed points to obtain a plurality of combined plurality of alternative points and a plurality of seed points; filtering the combined multiple candidate points and the multiple seed points through a voxel grid to obtain sampled seed points and candidate points; clustering the sampled candidate points and seed points by using a DBSCAN density clustering method, and presetting a minimum threshold value N of the point cloud number in the seed point cloud cluster s Acquiring a plurality of candidate point cloud clusters and seed point cloud clusters;
the central processing unit is further used for combining the seed point cloud cluster and the alternative point cloud cluster clustering result according to the minimum overlapping rate threshold k to obtain a seed point cloud cluster bounding box and an alternative point cloud cluster bounding box;
calculating bounding box volumes of the seed point cloud cluster and the candidate point cloud cluster respectively, judging whether the bounding box of the seed point cloud cluster is overlapped with the bounding box of the candidate point cloud cluster, and if so, calculating the overlapping rate of the bounding box of the seed point cloud cluster and the bounding box of the candidate point cloud cluster;
determining that the overlapping rate of the seed point cloud cluster bounding box and the alternative point cloud cluster bounding box is larger than the minimum overlapping rate threshold k, and determining that the alternative point cloud cluster corresponding to the alternative point cloud cluster bounding box is an effective tower point cloud cluster;
Calculating the superposition plane area of each effective tower point cloud cluster and each seed point cloud cluster, and determining the tower point cloud cluster with the largest superposition plane area as a target tower point cloud cluster;
the central processing unit is also used for traversing all the pole tower point clouds in the target pole tower point cloud cluster, calculating the pole tower bounding box volume and pole tower coordinates according to the coordinates of the pole tower point clouds, and extracting the highest point of the target pole tower point cloud cluster as the highest elevation of the pole tower center.
CN202310441901.3A 2023-04-23 2023-04-23 Robust unmanned plane laser radar line-imitating tower recognition method and device Pending CN116482700A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310441901.3A CN116482700A (en) 2023-04-23 2023-04-23 Robust unmanned plane laser radar line-imitating tower recognition method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310441901.3A CN116482700A (en) 2023-04-23 2023-04-23 Robust unmanned plane laser radar line-imitating tower recognition method and device

Publications (1)

Publication Number Publication Date
CN116482700A true CN116482700A (en) 2023-07-25

Family

ID=87222682

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310441901.3A Pending CN116482700A (en) 2023-04-23 2023-04-23 Robust unmanned plane laser radar line-imitating tower recognition method and device

Country Status (1)

Country Link
CN (1) CN116482700A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117115491A (en) * 2023-08-18 2023-11-24 国网山东省电力公司临沂供电公司 Method, system and storage medium for extracting protection angle of lightning conductor of power transmission tower pole based on laser point cloud data

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117115491A (en) * 2023-08-18 2023-11-24 国网山东省电力公司临沂供电公司 Method, system and storage medium for extracting protection angle of lightning conductor of power transmission tower pole based on laser point cloud data
CN117115491B (en) * 2023-08-18 2024-04-09 国网山东省电力公司临沂供电公司 Method, system and storage medium for extracting protection angle of lightning conductor of power transmission tower pole based on laser point cloud data

Similar Documents

Publication Publication Date Title
CN110781827A (en) Road edge detection system and method based on laser radar and fan-shaped space division
CA2839165C (en) Method and system for processing image data
CN110208819A (en) A kind of processing method of multiple barrier three-dimensional laser radar data
CN109816664B (en) Three-dimensional point cloud segmentation method and device
CN112883878B (en) Automatic point cloud classification method under transformer substation scene based on three-dimensional grid
US11295130B2 (en) Aerial line extraction system and aerial line extraction method
CN111325138B (en) Road boundary real-time detection method based on point cloud local concave-convex characteristics
CN109961470B (en) Living standing tree She Shuxing accurate estimation method based on laser point cloud
CN106845324B (en) Method and device for processing guideboard information
CN108133191A (en) A kind of real-time object identification method suitable for indoor environment
CN108562885B (en) High-voltage transmission line airborne LiDAR point cloud extraction method
CN110969142B (en) Abnormal driving scene extraction method based on network-connected vehicle natural driving data
CN111462134A (en) Single-tree segmentation method and system for fusing high-resolution remote sensing image and laser radar point cloud
CN107679458B (en) Method for extracting road marking lines in road color laser point cloud based on K-Means
CN113345094A (en) Electric power corridor safety distance analysis method and system based on three-dimensional point cloud
CN116482700A (en) Robust unmanned plane laser radar line-imitating tower recognition method and device
CN111783721A (en) Lane line extraction method of laser point cloud and electronic equipment
CN112381041A (en) Tree identification method and device for power transmission line and terminal equipment
CN113469178A (en) Electric power meter identification method based on deep learning
CN115825984A (en) Power transmission line inspection system and inspection method based on laser point cloud
CN114782947A (en) Point cloud matching method, point cloud matching system and storage medium for power transmission and distribution line
CN114639023A (en) Roof photovoltaic potential evaluation method based on machine learning
CN117826854A (en) Electric power inspection route fully-autonomous planning method based on three-dimensional space target point extraction
CN116893685A (en) Unmanned aerial vehicle route planning method and system
CN117152188A (en) Point cloud data-based power transmission line insulator extraction method

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