CN114859368A - Method and system for tracking and processing power line locking by using laser radar - Google Patents
Method and system for tracking and processing power line locking by using laser radar Download PDFInfo
- Publication number
- CN114859368A CN114859368A CN202210475307.1A CN202210475307A CN114859368A CN 114859368 A CN114859368 A CN 114859368A CN 202210475307 A CN202210475307 A CN 202210475307A CN 114859368 A CN114859368 A CN 114859368A
- Authority
- CN
- China
- Prior art keywords
- line
- straight line
- point
- straight
- tower
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/66—Tracking systems using electromagnetic waves other than radio waves
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
-
- 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
Abstract
The invention discloses a method and a system for tracking and processing power line locking by using a laser radar, wherein the method comprises the steps of acquiring laser point cloud in real time by using the laser radar to obtain point cloud data; dividing the laser point cloud by using a deep learning model to obtain a power line point and a tower point; clustering the tower points, and calculating to generate the center point coordinates of the tower points; extracting all straight lines of a current line channel by using an RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all the straight lines; selecting a target tracking straight line; judging whether the deviation angle and the distance of the target tracking straight line are smaller than or equal to a preset standard deviation threshold value or not; and if so, predicting the flight prediction position of the aircraft by using the target tracking straight line. The technical scheme of the invention can solve the problems of unstable locking line tracking, low traditional hough transformation efficiency, large occupation of the memory and high requirements on the computing capacity and computing resources of the computing unit in the prior art.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a system for tracking and processing power line locking by using a laser radar.
Background
The related art of spatial ranging often needs to measure and extract straight lines in three-dimensional space. Such as mapping, drone reconnaissance and transmission line measurement techniques. In the existing transmission line measurement, for measurement and extraction of straight lines, generally, an unmanned aerial vehicle patrols and examines to obtain relevant images of a general transmission tower, and then a relevant unmanned aerial vehicle line tracing algorithm is used for extracting and obtaining straight line data in a three-dimensional space.
The scheme that the camera shoots images is mostly adopted in the related technology related to the current unmanned aerial vehicle line tracing algorithm, specifically, the line is shot, and then a hough transformation algorithm is adopted to extract a line equation. The defects of the scheme are that the camera is easily influenced by illumination and needs to shoot in a close range, otherwise, the resolution is not enough or the image is not clear, the line in the image cannot be identified, in addition, the image shooting line is shielded, and a background straight line in a scene is also easily extracted, and the factors enable the scheme for shooting the image by the camera to be unstable and reliable.
Disclosure of Invention
The invention provides a method for tracking a power line lock by using a laser radar, and a linear equation of all straight lines on a current line channel is extracted by using a RANSAC multi-line extraction algorithm with constraint conditions, aiming at solving the problems of unstable lock line tracking, low efficiency of the traditional hough transformation and least square algorithm, large occupation of a memory, and high requirements on computing capacity and computing resources of a computing unit in the prior art.
In order to achieve the above object, according to a first aspect of the present invention, the present invention provides a method for tracking and processing power line lockwire by using a laser radar, the method mainly includes:
collecting laser point cloud in real time by using a laser radar to obtain point cloud data;
dividing the laser point cloud contained in the point cloud data by using a deep learning model to obtain a power line point and a tower point corresponding to the current line channel;
clustering pole tower points in the current line channel, and calculating and generating center point coordinates of the pole tower points by using the clustered pole tower points;
extracting all straight lines of the current line channel according to the center point coordinates of the pole and tower points by using a RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all straight lines in the current line channel;
selecting a target tracking straight line according to the straight line equations of all straight lines in the current line channel;
judging whether the deviation angle and the distance of the target tracking straight line are smaller than or equal to a preset standard deviation threshold value or not;
and if the deviation angle and the distance are judged to be smaller than or equal to the preset standard deviation threshold value, the target tracking straight line is used for predicting the flight prediction position of the aircraft.
Preferably, the step of clustering the tower points in the current line channel and calculating and generating the coordinates of the center point of the tower points by using the clustered tower points includes:
extracting tower points located within a preset distance range of a current line channel;
clustering the pole and tower points by using a DBSCAN algorithm, and selecting the category with the most pole and tower points;
and averaging and calculating coordinates of all tower points in the category with the largest number of tower points to obtain the coordinate of the center point of the tower point.
Preferably, in the method for tracking locked line of power line, a RANSAC multiline extraction algorithm with constraint conditions is used to extract all straight lines of the current line channel according to the coordinates of the center point of the tower point, and the method includes:
extracting circuit line points corresponding to a current line channel by using an RANSAC multi-line extraction algorithm to obtain a current frame straight line;
acquiring a previous frame straight line of the current frame straight line, and constraining the current frame straight line according to the direction of the previous frame straight line so as to judge whether the included angle of the two frame straight lines exceeds a preset angle threshold value;
if the included angle of the two straight lines exceeds the preset angle threshold, abandoning the current straight line and moving the power line point out of the current line channel;
if the included angle between the two frame straight lines does not exceed the preset angle threshold, determining the straight line section corresponding to the current frame straight line as an effective line section;
acquiring the center coordinate of the nearest tower point, and constraining the effective line segment according to the position relation between the center coordinate of the nearest tower point and the effective line segment;
constraining the effective line segment according to the distance between the effective line segment and a power line point of the current line channel;
and extracting the restrained effective line segments and repeating the steps until all the straight lines are extracted.
Preferably, in the power line locking tracking processing method, the step of obtaining the power line point and the tower point corresponding to the current line channel includes:
acquiring the current position and the flight direction of the aircraft;
selecting a line channel which is closest to the current position and the flight direction of the aircraft in the three-dimensional space;
and selecting a power line point and a tower point corresponding to the current line channel from the point cloud data.
Preferably, in the above power line lock tracking processing method, according to a preset voting score mechanism, using a linear equation of all the straight lines in the current line channel, and selecting the straight line with the highest voting score as the target tracking straight line includes:
calculating the distance from each straight line to the eccentric coordinate by using the linear equation of all straight lines in the current line channel;
voting scoring is carried out on the distance from each straight line to the eccentric coordinate according to a preset voting mechanism, and the straight line with the highest voting scoring is selected as a target tracking straight line;
and constraining the target tracking straight line to obtain a straight line equation of the target tracking straight line.
Preferably, in the power line lock tracking processing method, the step of calculating the distance from each straight line to the eccentric coordinate includes:
calculating the coordinates of the central point of the line channel by using the initial coordinates of all straight lines in the line channel;
calculating an eccentric coordinate by using the point coordinates from the center point coordinate to the upper closest point of all the straight lines;
the distance of each straight line to the eccentric coordinates is calculated.
Preferably, the voting scoring by the preset voting scoring mechanism for the distance between each straight line and the eccentric coordinate includes:
substituting the three-dimensional distance, the height and the average distance from each straight line to the eccentric coordinate into a straight line score formula corresponding to the preset voting score mechanism to obtain a score corresponding to each straight line;
and selecting the straight line with the highest score as a target tracking straight line.
Preferably, in the power line lock tracking processing method, the step of constraining the target tracking straight line to obtain a straight line equation of the target tracking straight line includes:
selecting a target tracking straight line in real time by using a preset line priority principle to obtain a multi-frame straight line corresponding to the target tracking straight line;
and using the straight line of the previous frame in the multi-frame straight lines to constrain the straight line of the current frame to obtain a straight line equation of the straight line of the current frame, and using the straight line equation as a straight line equation of the target tracking straight line.
According to a second aspect of the present invention, the present invention further provides a system for tracking and processing power line locking by using a laser radar, including:
the point cloud acquisition module is used for acquiring laser point cloud in real time by using a laser radar to obtain point cloud data;
the point cloud segmentation module is used for segmenting laser point cloud contained in the point cloud data by using a deep learning model to obtain a power line point and a tower point corresponding to a current line channel;
the extraction clustering module is used for clustering the tower points in the current line channel and calculating and generating the central point coordinates of the tower points by using the clustered tower points;
the straight line extraction module is used for extracting all straight lines of the current line channel according to the center point coordinates of the pole and tower points by using a RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all straight lines in the current line channel;
the target tracking straight line selection module is used for selecting a target tracking straight line according to the straight line equations of all straight lines in the current line channel;
the deviation angle judging module is used for judging whether the deviation angle of the target tracking straight line is smaller than or equal to a preset standard deviation threshold value or not;
and the flight prediction module is used for predicting the flight prediction position of the aircraft by using the target tracking straight line if the deviation angle is judged to be less than or equal to the preset standard deviation threshold value.
According to a third aspect of the present invention, the present invention further provides a system for tracking and processing power line locking by using a laser radar, including:
the system comprises a memory, a processor and a program which is stored on the memory and can run on the processor and is used for tracking and processing the power line locking by using the laser radar, wherein when the program is executed by the processor, the steps of the method for tracking and processing the power line locking by using the laser radar in any technical scheme are realized.
In summary, according to the power line locking tracking processing scheme provided by the technical scheme of the application, the laser point cloud is collected in real time by using the laser radar to obtain the point cloud data, and the laser point cloud contained in the point cloud data is segmented by using the deep learning model to obtain the power line point and the tower point corresponding to the line channel. Then extracting and clustering tower points in the current line channel, generating central point coordinates of the tower points by using the clustered tower points, and then performing straight line extraction on the central point coordinates of the tower points by using a RANSAC multi-line extraction algorithm with constraint conditions, so that a target tracking straight line in the line channel can be obtained. The flight prediction position of the aircraft can be accurately predicted by using the target tracking straight line. According to the technical scheme, the point cloud data are obtained through the laser radar, so that the power line point and the tower point corresponding to the current line channel are obtained, and stable line locking tracking can be guaranteed. In addition, the coordinates of the central point of the pole and tower point are sampled by using a random sampling consistency RANSAC multiline extraction algorithm to extract a straight line in a line channel, the random sampling consistency RANSAC multiline extraction algorithm has constraint conditions and can quickly and accurately extract the straight line, and the RANSAC multiline extraction algorithm has higher efficiency and smaller memory occupation compared with a hough transformation algorithm, so that the problem that the hough algorithm in the prior art has higher requirements on the computing capacity and computing resources of a computing unit can be solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a first method for tracking a power line lock using a lidar according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for selecting a circuit path according to the embodiment shown in FIG. 1;
FIG. 3 is a schematic flow chart of a tower point extraction and clustering method provided by the embodiment shown in FIG. 1;
fig. 4 is a schematic flowchart of a method for extracting a straight line from coordinates of a center point of a tower point according to the embodiment shown in fig. 1;
fig. 5 is a schematic flowchart of a second method for tracking a power line lock using a lidar according to an embodiment of the present invention;
FIG. 6 is a schematic flow chart of a method for selecting a target tracking line according to the embodiment shown in FIG. 5;
FIG. 7 is a flow chart illustrating a straight line scoring method provided by the embodiment shown in FIG. 6;
FIG. 8 is a flow chart illustrating a method for obtaining the equation of a straight line according to the embodiment shown in FIG. 5;
fig. 9 is a schematic structural diagram of a first system for tracking and processing locked power lines based on a RANSAC multiline extraction algorithm according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of a second system for tracking locked line of power line based on RANSAC multiline extraction algorithm according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The main solution of the embodiment of the invention is as follows:
most of the existing straight line extraction algorithms adopt hough transformation algorithms. However, the hough transform algorithm has low efficiency, occupies a large memory, and has high requirements on the computing capacity and computing resources of the computing unit, which limits the application range of hough. For example, the computing unit carried by the unmanned aerial vehicle platform has relatively weak computing power and precious computing resources, and deployment of the unmanned aerial vehicle platform is not facilitated by adopting hough transformation.
In order to solve the above problems, the following embodiments of the present invention provide a power line lock tracking processing scheme based on a RANSAC multiline extraction algorithm, where a random sampling consistency RANSAC multiline extraction algorithm is used to sample power line points in a line channel, so as to extract straight lines in the line channel, the extraction efficiency is high, and random sampling occupies a small memory, so that the problem that a hough algorithm in the prior art has high requirements on the computing power and computing resources of a computing unit can be solved.
To achieve the above object, please refer to fig. 1, where fig. 1 is a schematic flow chart of a first laser radar power line locking tracking processing method according to an embodiment of the present invention. As shown in fig. 1, the power line locking tracking processing method includes:
s110: and collecting laser point cloud in real time by using a laser radar to obtain point cloud data. The laser radar is used for collecting the point cloud, the point cloud can not be influenced by illumination, long-distance point cloud collection can be realized, and the problems that an existing camera is easy to influence the collected image by illumination and needs to shoot in a close range, and otherwise, the line cannot be identified insufficiently or the image is not clear and the line in the image cannot be identified are solved.
S120: and dividing the laser point cloud contained in the point cloud data by using a deep learning model to obtain a power line point and a tower point corresponding to the current line channel.
Because a deeply learned and predicted line channel (for example, a line channel of a power line) often has some noises, and a plurality of lines in a scene may be completely predicted and extracted, a line of the line channel needs to be tracked and locked, and a line tracing initialization is needed at this time. Specifically, the general direction of the flight of the unmanned aerial vehicle is determined, the line (such as a power line) closest to the position and the direction of the unmanned aerial vehicle at present is selected, the same line channel is locked twice continuously, then the initialization is successful, and the flight direction and the line channel are determined.
Specifically, as a preferred embodiment, as shown in fig. 2, the step of obtaining the power line point and the tower point corresponding to the current line channel includes:
s121: acquiring the current position and the flight direction of the aircraft;
s122: selecting a line channel which is closest to the current position and the flight direction of the aircraft in the three-dimensional space as a current line channel;
s123: and selecting a power line point and a tower point corresponding to the current line channel from the point cloud data.
By selecting the route on which the current position of the aircraft is closest to the flight direction, for example, which passes through the coordinates of the current position of the aircraft and which does not include a predetermined angle threshold with respect to the flight direction, the same route channel is determined when the same route channel is locked twice in succession.
In addition, after selecting a line channel in a three-dimensional space, the laser radar power line locking tracking processing method shown in fig. 1 further includes:
s130: and clustering the tower points in the current line channel, and calculating to generate the central point coordinates of the tower points by using the clustered tower points.
Taking the power line channel as an example, the embodiment of the application can extract the power line channels of multiple frames, each power line channel of the multiple frames comprises multiple tower points, so that the power line channel extracted from the previous frame can be used for constraining the power line channel of the next frame, and the coordinates of almost all the tower points on the power line channel are obtained. After a large number of coordinates of the tower points are obtained, the tower points can be clustered, and the embodiment of the application uses a DBSCAN algorithm to cluster the tower points. After clustering is successful, coordinates of a large number of tower points are used, and center point coordinates of the tower points can be generated according to a certain algorithm (such as weighting averaging or selecting the tower point with the coordinate closest to the average value).
Specifically, as a preferred embodiment, as shown in fig. 3, the step of extracting and clustering tower points in the line channel, and using the clustered tower point coordinates to generate center point coordinates of the tower points includes:
s131: and extracting tower points located in a preset distance range of the current line channel, wherein the tower points are tower points detected and output by the line channel through a deep learning model.
S132: and clustering the pole and tower points by using a DBSCAN algorithm, and selecting the class with the most pole and tower points.
S133: and averaging and calculating coordinates of all tower points in the category with the largest number of tower points to obtain the coordinate of the center point of the tower point.
Among them, DBSCAN (sensitivity-Based Spatial Clustering of applications with noise) is a relatively representative Density-Based Clustering algorithm. Unlike the partitioning and hierarchical clustering method, which defines clusters as the largest set of density-connected points, it is possible to partition areas with sufficiently high density into clusters and find clusters of arbitrary shape in a spatial database of noise.
Taking tower point extraction and clustering as an example, the embodiment of the application utilizes the power line channel extracted from the previous frame to constrain the tower points output by each frame of the deep learning model in the future, and tower points exceeding a certain distance range are filtered out, and points of the current line are left. And then accumulating the rest tower points, clustering by adopting a DBSCAN algorithm, selecting the largest class as the class of the tower points, confirming that the tower is identified and effective when the number of the clustered points reaches a certain threshold value, and calculating the coordinates of the center point of the tower by using the coordinates of the clustered points.
After the clustered tower points are used to generate center point coordinates of the tower points, the method for tracking the locked lines of the power line provided by the embodiment shown in fig. 1 further includes the following steps:
s140: and extracting all straight lines of the current line channel according to the central point coordinates of the tower point by using a RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all straight lines in the current line channel.
RANSAC multiline extraction algorithm with constraint conditions:
the RANSAC multi-line extraction algorithm with random sampling consistency is an algorithm for calculating mathematical model parameters of data according to a group of sample data sets containing abnormal data to obtain effective sample data. The algorithm has the characteristics of high sampling efficiency and small occupied resource. The RANSAC multi-line extraction algorithm is used for extracting straight lines of power line points (including partial noise points) output by the deep learning model to obtain straight lines in a three-dimensional space scene, and a corresponding straight line equation can be generated according to the straight lines. Taking the power channel as an example, because the cloud point in the power channel may have noise points and points on two sides of the tower are on different straight lines when passing through the tower, the points which are not on the same straight line are easily regarded as the same straight line by adopting the traditional RANSAC multi-line extraction algorithm, and certain errors exist. In addition, the classical RANSAC multiline extraction algorithm tends to easily extract the same straight line in a line into a plurality of straight lines. Therefore, the embodiment of the application provides a RANSAC multi-line extraction algorithm with constraint conditions, which utilizes the straight line of the previous frame and the coordinates of the tower to constrain the current straight line, and simultaneously screens the straight lines in the extraction process, so as to prevent the same straight line from being detected as a plurality of straight lines, and finally realize accurate extraction of the straight lines of the line.
Specifically, as a preferred embodiment, as shown in fig. 4, the step of performing line extraction using the RANSAC multiline extraction algorithm with constraint conditions includes:
s141: extracting power line points corresponding to the current line channel by using the RANSAC multi-line extraction algorithm to obtain a current frame straight line;
s142: if the included angle of the two straight lines exceeds the preset angle threshold, abandoning the current straight line and moving the power line point out of the current line channel;
s143: if the included angle between the two frame straight lines does not exceed the preset angle threshold, determining the straight line section corresponding to the current frame straight line as an effective line section;
s144: acquiring the center coordinate of the nearest tower point, and constraining the effective line segment according to the position relation between the center coordinate of the nearest tower point and the effective line segment;
s145: constraining the effective line segment according to the distance between the effective line segment and a power line point of the current line channel;
s146: and extracting the restrained effective line segments and repeating the steps until all the straight lines are extracted.
In summary, the step of extracting all the straight lines of the current line channel according to the coordinates of the center point of the tower point is specifically as follows:
1. extracting a straight line from original power line points (including partial noise points) given by a model by using a traditional RANSAC multi-line extraction algorithm;
2. and (3) constraining the current straight line by using the direction of the previous frame straight line (if the previous frame straight line does not exist, the constraint is not carried out), if the included angle of the two straight lines exceeds a threshold value, the current straight line is considered to have large deviation and is possibly greatly influenced by noise points, the straight line is abandoned, the corresponding point is taken as an external point and removed from the original power line point, RANSAC straight line extraction is not carried out, and the step 1 is repeated. If the included angle is not over-limit, the straight line segment is an effective line segment.
3. Calculating a point closest to a tower on a linear equation, if a tower point is located in the middle of a linear segment, the current line is actually two lines, the linear segment needs to be divided into two sections by using the tower, one of the two sections is selected, the points of the remaining one section are placed into the original power line point for further extraction, if the tower is not located in the middle of the linear segment, but located far from the two sides, the current linear is regarded as actually only one line, and no additional processing is needed.
4. Setting a threshold value d by using a current straight line equation, calculating the distance from an original power line point to a current straight line, if the distance is less than the distance d, considering a point position straight line close point or a point on the straight line, completing segment segmentation by using the step 3 to obtain a point near the current straight line, removing the point near the current straight line from the original power line point, and not participating in straight line extraction any more, thereby preventing a single straight line from being extracted into a plurality of straight lines.
5. And saving the current straight line, and repeating the steps until all the straight lines are extracted.
Remarking: the above is the RANSAC multiline extraction algorithm process. Corresponding calculation formulas and calculation methods can be provided if necessary. (e.g., calculation of the nearest point of a straight line segment, calculation of the included angle of a straight line, calculation of the distance from a point to a line, etc.).
In the technical scheme provided by the embodiment of the application, in the process of randomly selecting points through RANSAC and continuously adding points, a linear equation and the nearest coordinate P of the tower point obtained in the steps on the linear equation are calculated, whether the coordinate P is in the middle of a linear segment behind the added tower point is judged, if the coordinate P is in the middle of the linear segment, a point which is subsequently added and is not on an effective segment (the range from the starting point position of the unmanned aerial vehicle to the position of the tower is an effective range) is not taken as a linear point, and otherwise, the point is added into the linear and is taken as an inner point of the linear.
And then, the direction of the straight line of the previous frame is utilized to restrain the extraction of the current straight line, if the included angle of the two straight lines does not exceed a certain angle threshold value, when the power line point is randomly selected by the current frame for straight line extraction, the power line point falling on the straight line is selected to be added into the straight line as an inner point, otherwise, the straight line is not added, and great errors caused by excessive noise points of the channel points are prevented.
Because a plurality of straight lines often exist in a line channel, such as a power line channel, after a straight line is extracted each time, straight line points corresponding to the straight line are removed, the remaining points are remained, the above processes are continuously repeated, and the straight lines are extracted again until all the straight lines are extracted.
S150: and according to a preset voting score mechanism, using the linear equations of all the straight lines in the current line channel, and selecting the straight line with the highest voting score as the target tracking straight line.
S160: judging whether the deviation angle and the distance of the target tracking straight line are smaller than or equal to a preset standard deviation threshold value or not; if yes, go to step S170; if not, the process returns to step S110, and starts the process of the next frame line.
S170: and if the deviation angle and the distance are judged to be smaller than or equal to the preset standard deviation threshold value, the target tracking straight line is used for predicting the flight prediction position of the aircraft.
The prediction method of the flight prediction position, namely the flight track point, comprises the following specific steps:
in order to realize autonomous flight of the unmanned aerial vehicle, after an optimal line linear equation is extracted by using the laser radar point cloud of the unmanned aerial vehicle, the position of the next flight can be calculated according to the linear equation. The expression of the space linear equation is:
X＝n x ·t+X 0
Y＝n y ·t+Y 0
Z＝n z ·t+Z 0
wherein (n) x ,n y ,n z ) Is a straight line direction (X) 0 ,Y 0 ,Z 0 ) At one point in the line, t is an arbitrary scalar.
Then, the position of the next point, i.e. the predicted position of flight, can be calculated according to the position coordinates of the current point of the airplane, the direction of the straight line and the predicted distance t.
In summary, the power line locking tracking processing method provided by the above technical solution of the present application selects a line channel in a three-dimensional space, then extracting and clustering pole tower points in the line channel, generating central point coordinates of the pole tower points by using the clustered pole tower points, then using RANSAC multiline extraction algorithm with constraint conditions to extract straight lines of the power line points in the channel, obtaining straight lines in the line channel, as the technical scheme of the application uses the RANSAC multi-line extraction algorithm with random sampling consistency to sample the power line points in the channel so as to extract straight lines in the channel, moreover, the method has constraint conditions, can quickly and accurately extract straight lines, has higher efficiency of the RANSAC multiline extraction algorithm compared with hough transformation algorithm, the occupied internal memory is small, so that the problem that the hough algorithm in the prior art has high requirements on the computing capacity and computing resources of a computing unit can be solved.
Because the line channel comprises a large number of straight lines, after the straight lines of the line channel are extracted, the straight line with the minimum deviation angle and distance can be selected for aircraft navigation application.
Specifically, as a preferred embodiment, as shown in fig. 5, the power line lockwire tracking processing method provided in the embodiment of the present application, where the step of selecting, according to a preset voting score mechanism, a straight line equation of all straight lines in a current line channel and using a straight line with a highest voting score as a target tracking straight line specifically includes:
s210: and calculating the distance from each straight line to the eccentric coordinate by using the linear equation of all straight lines in the current line channel.
As shown in fig. 6, the step of calculating the distance from each straight line to the eccentric coordinate by using the equation of the straight line of all the straight lines in the current line channel includes:
s211: filtering the straight line in the line channel according to the position information of the line channel so as to remove the straight line in the non-line channel;
s212: calculating the coordinates of the central point of the line channel by using the initial coordinates of all straight lines in the line channel;
s213: calculating an eccentric coordinate by using the point coordinates from the center point coordinate to the upper closest point of all the straight lines;
s214: the distances of all straight lines to the eccentric coordinates are calculated.
The embodiment of the application needs to filter straight lines in a scene, and the straight lines extracted in a general case are channels of all present ground features in a ground scene, including channels in parallel lines, for example. And filtering straight lines in other unrelated channels, and carrying out constraint by using information such as the distance and the angle of the line channel locked by the aircraft to remove lines which are not in the current channel so as to obtain all straight lines in the current line.
After all the straight lines in the current line are obtained, a specific line needs to be selected and locked. Because there are multiple straight lines in one channel, the same line needs to be locked during flight of the aircraft to keep the flight smooth.
Therefore, after calculating the distance from each straight line to the eccentric coordinate, the method shown in fig. 5 further includes the following steps:
s220: and voting and scoring the distance from each straight line to the eccentric coordinate according to a preset voting and scoring mechanism, and selecting the straight line with the highest voting and scoring as a target tracking straight line.
Specifically, as a preferred embodiment, as shown in fig. 7, the step of scoring the distance from each straight line to the eccentric coordinate according to the preset voting score mechanism includes:
s221: substituting the three-dimensional distance, the height and the average distance from each straight line to the eccentric coordinate into a straight line score formula corresponding to the preset voting score mechanism to obtain a score corresponding to each straight line;
s222: and selecting the straight line with the highest score as the target tracking straight line.
The specific route selection and locking tracking algorithm process is as follows:
the unmanned aerial vehicle is in flight in-process real-time tracking line need lock the tracking to same power line in the passageway, otherwise in many circuits switching process, will lead to the flight position of prediction to produce big skew, finally makes the flight unstable. Therefore, a locking line tracking method based on a preset voting score mechanism is provided, the score of each line is calculated, and the line with the highest score is used as the current frame optimal line.
The detailed process of the lockwire tracking algorithm is as follows:
and calculating the coordinate Pc of the average center point according to the coordinates of the starting point and the ending point of all the straight line segments in the line channel.
And calculating the coordinates Pf of the point from the center point coordinates to the nearest point on all the straight lines by using the center point coordinates Pc.
And calculating azimuth angles by using the center point coordinate Pc and the direction from the center point coordinate to the nearest point coordinate Pf of the straight line, obtaining the azimuth angle from the center point coordinate to each straight line, and sequencing all the azimuth angles from small to large.
After the sorting in the previous step, the coordinates Pf of the closest point of the first half are averaged to obtain the eccentric coordinates Pe, at the moment, the line division is completed, and only one side of the line on the left side and the right side of the channel center is selected.
The 3d distances d of all the straight lines to the point eccentric coordinates Pe are calculated and dmin and dmax are obtained. The score Sc for each line is calculated as (1- (d-dmin)/(dmax-dmin)). 100, with lines closer to the center of the channel having higher scores and lines closer to the center having lower scores, i.e., preferentially tracing lines closer to the center.
The height H of all the straight lines at the eccentric coordinates Pe is calculated, and Hmin and Hmax are obtained. The score Sh of all lines is calculated as (H-Hmin)/(Hmax-Hmin) × 100, with higher lines in the line scoring higher and vice versa, i.e. higher lines are tracked preferentially.
The distances from all the straight lines to the previous frame straight line (if any) at the eccentric coordinate Pe and at N points equidistant from the coordinate Pe along the straight line direction are calculated, and the distances D (i.e. the average distance) are obtained by averaging, and Dmin and Dmax are obtained. And calculating the score Sr of each straight line as (1- (D-Dmin)/(Dmax-Dmin)). 100, wherein the score of the straight line segment which is closer to the previous frame in the line is higher, and the score of the straight line segment which is closer to the previous frame in the line is lower, namely the straight line which is closest to the straight line in the previous frame is preferentially tracked, and the process realizes the tracking of the same straight line.
And calculating the score S of each straight line to be Sc + Sh + Sr, and selecting the straight line with the highest score as the line needing to be locked in flight.
Calculating the deviation between the current straight line and the straight line of the previous frame, calculating the included angle of the two straight lines, sampling at equal intervals in the range of the line segment, and calculating the direct distance between the two straight lines, wherein if the included angle and the distance exceed the range of a threshold value, the current best straight line is considered to be not available, and if not, the best straight line is obtained for next step of track prediction.
Remarking: the above process is a lockwire tracking algorithm.
After the target tracking straight line is selected as the target tracking straight line of the aircraft, the method for extracting the straight line in the three-dimensional space shown in fig. 5 further includes the following steps:
s230: and constraining the target tracking straight line to obtain a straight line equation of the target tracking straight line.
As a preferred embodiment, as shown in fig. 8, the step of constraining the target tracking straight line to obtain the straight line equation of the target tracking straight line includes:
s231: selecting a target tracking straight line in real time by using a preset line priority principle to obtain a multi-frame straight line corresponding to the target tracking straight line;
s232: and using the previous frame straight line to constrain the current frame straight line to obtain a straight line equation of the current frame straight line as a straight line equation of the target tracking straight line.
The line selection method and the line selection device adopt the preset line priority principle that the height is high and the left or right line is always preferred. And after the initialization is successful, the line of the current frame is locked by utilizing the straight line constraint of the previous frame to obtain the straight line equation of the current frame.
Based on the same concept of the embodiment of the method, the embodiment of the invention also provides a system for tracking and processing the locked line of the power line based on the RANSAC multi-line extraction algorithm, which is used for realizing the method of the invention.
Referring to fig. 9, fig. 9 is a schematic structural diagram of a power line lock tracking processing system based on a RANSAC multiline extraction algorithm according to an embodiment of the present invention. As shown in fig. 9, the system for tracking and processing power line lock using lidar is characterized by comprising:
a point cloud obtaining module 110, configured to collect a laser point cloud in real time by using a laser radar, so as to obtain the point cloud data;
the point cloud segmentation module 120 is configured to segment the laser point cloud included in the point cloud data by using a deep learning model, and obtain a power line point and a tower point corresponding to a current line channel;
an extracting and clustering module 130, configured to cluster the tower points in the current line channel, and calculate and generate a central point coordinate of the tower point by using the clustered tower points;
the straight line extraction module 140 is configured to extract all straight lines of the current line channel according to the center point coordinates of the tower point by using a RANSAC multi-line extraction algorithm with a constraint condition, so as to obtain straight line equations of all straight lines in the current line channel;
a target tracking straight line selection module 150, configured to select a target tracking straight line according to the straight line equations of all straight lines in the current line channel;
a deviation angle determination module 160, configured to determine whether a deviation angle of the target tracking straight line is smaller than or equal to a preset standard deviation threshold;
and the flight prediction module 170 is configured to predict a flight prediction position of the aircraft by using the target tracking straight line if it is determined that the deviation angle is smaller than or equal to the preset standard deviation threshold.
The power line locking tracking processing system provided by the embodiment of the application selects the line channel in the three-dimensional space, then extracting and clustering pole tower points in the line channel, generating central point coordinates of the pole tower points by using the clustered pole tower points, then using RANSAC multiline extraction algorithm with constraint conditions to perform straight line extraction on the power line points predicted by the deep learning model, so as to obtain straight lines in a line channel, as the technical scheme of the application uses the RANSAC multiline extraction algorithm with random sampling consistency to sample the power line points so as to extract straight lines in the line channel, moreover, the method has constraint conditions, can quickly and accurately extract straight lines, has higher efficiency of the RANSAC multiline extraction algorithm compared with hough transformation algorithm, the occupied internal memory is small, so that the problem that the hough algorithm in the prior art has high requirements on the computing capacity and computing resources of a computing unit can be solved.
In addition, as shown in fig. 10, an embodiment of the present invention further provides a schematic structural diagram of a power line locking tracking processing system. This power line locking tracking processing system includes:
the system comprises a communication line 1002, a communication module 1003, a memory 1004, a processor 1001 and a power line locking tracking processing program which is stored on the memory 1004 and can be run on the processor 1001, wherein when the power line locking tracking processing program is executed by the processor 1001, the steps of the power line locking tracking processing method provided by any one of the technical schemes are realized.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A method for tracking and processing power line locking by using a laser radar is characterized by comprising the following steps:
collecting laser point cloud in real time by using a laser radar to obtain point cloud data;
dividing the laser point cloud contained in the point cloud data by using a deep learning model to obtain a power line point and a tower point corresponding to the current line channel;
clustering pole tower points in the current line channel, and calculating and generating center point coordinates of the pole tower points by using the clustered pole tower points;
extracting all straight lines of the current line channel according to the center point coordinates of the pole and tower points by using a RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all straight lines in the current line channel;
according to a preset voting score mechanism, using the linear equations of all the straight lines in the current line channel, and selecting the straight line with the highest voting score as a target tracking straight line;
judging whether the deviation angle and the distance of the target tracking straight line are smaller than or equal to a preset standard deviation threshold value or not;
and if the deviation angle and the distance are judged to be smaller than or equal to the preset standard deviation threshold value, the target tracking straight line is used for predicting the flight prediction position of the aircraft.
2. The method for tracking and processing the locking of the power line according to claim 1, wherein the step of clustering the tower points in the current line channel and calculating and generating the coordinates of the center point of the tower points by using the clustered tower points comprises:
extracting tower points located in a preset distance range of the current line channel;
clustering the pole and tower points by using a DBSCAN algorithm, and selecting the category with the largest pole and tower points;
and averaging and calculating coordinates of all tower points in the category with the maximum tower points to obtain the coordinate of the central point of the tower point.
3. The method for tracking and processing the locked power line according to claim 1 or 2, wherein the step of extracting all the straight lines of the current line channel according to the coordinates of the center point of the tower point by using RANSAC multiline extraction algorithm with constraint conditions comprises:
extracting power line points corresponding to the current line channel by using the RANSAC multi-line extraction algorithm to obtain a current frame straight line;
acquiring a previous frame straight line of the current frame straight line, and constraining the current frame straight line according to the direction of the previous frame straight line so as to judge whether the included angle of the two frame straight lines exceeds a preset angle threshold value;
if the included angle of the two straight lines exceeds the preset angle threshold, abandoning the current straight line and moving the power line point out of the current line channel;
if the included angle between the two frame straight lines does not exceed the preset angle threshold, determining the straight line section corresponding to the current frame straight line as an effective line section;
acquiring the center coordinate of the nearest tower point, and constraining the effective line segment according to the position relation between the center coordinate of the nearest tower point and the effective line segment;
constraining the effective line segment according to the distance between the effective line segment and a power line point of the current line channel;
and extracting the restrained effective line segments and repeating the steps until all the straight lines are extracted.
4. The method for tracking and processing the locking of the power line according to claim 1, wherein the step of obtaining the power line point and the tower point corresponding to the current line channel comprises:
acquiring the current position and the flight direction of the aircraft;
selecting a line channel which is closest to the current position and the flight direction of the aircraft in the three-dimensional space;
and selecting a power line point and a tower point corresponding to the current line channel from the point cloud data.
5. The method for tracking and processing the power line lock according to claim 1 or 4, wherein the step of selecting the straight line with the highest voting score as the target tracking straight line by using the straight line equations of all the straight lines in the current line channel according to a preset voting score mechanism comprises:
calculating the distance from each straight line to the eccentric coordinate by using the linear equation of all straight lines in the current line channel;
voting scoring is carried out on the distance from each straight line to the eccentric coordinate according to the preset voting mechanism, and the straight line with the highest voting score is selected as the target tracking straight line;
and constraining the target tracking straight line to obtain a straight line equation of the target tracking straight line.
6. The method for tracking and processing the locking of the power line according to claim 5, wherein the step of calculating the distance from each straight line to the eccentric coordinate comprises:
calculating the coordinates of the center point of the line channel by using the initial coordinates of all straight lines in the line channel;
calculating the eccentric coordinates by using the point coordinates from the central point coordinates to the nearest points on all the straight lines;
and calculating the distance from each straight line to the eccentric coordinate.
7. The method for tracking and processing the power line lock according to claim 5, wherein the voting scoring of the distance from each straight line to the eccentric coordinate according to the preset voting scoring mechanism comprises:
substituting the three-dimensional distance, the height and the average distance from each straight line to the eccentric coordinate into a straight line score formula corresponding to the preset voting score mechanism to obtain a score corresponding to each straight line;
and selecting the straight line with the highest score as the target tracking straight line.
8. The method for tracking and processing the power line lock according to claim 5, wherein the step of constraining the target tracking straight line to obtain the straight line equation of the target tracking straight line comprises:
selecting the target tracking straight line in real time by using a preset line priority principle to obtain a multi-frame straight line corresponding to the target tracking straight line;
and using the straight line of the previous frame in the multi-frame straight lines to constrain the straight line of the current frame to obtain a straight line equation of the straight line of the current frame, and using the straight line equation as the straight line equation of the target tracking straight line.
9. A system for power line lockdown tracking processing using a lidar, comprising:
the point cloud acquisition module is used for acquiring laser point cloud in real time by using a laser radar to obtain point cloud data;
the point cloud segmentation module is used for segmenting laser point cloud contained in the point cloud data by using a deep learning model to obtain a power line point and a tower point corresponding to a current line channel;
the extraction clustering module is used for clustering the tower points in the current line channel and calculating and generating the central point coordinates of the tower points by using the clustered tower points;
the straight line extraction module is used for extracting all straight lines of the current line channel according to the center point coordinates of the pole and tower points by using a RANSAC multi-line extraction algorithm with constraint conditions to obtain straight line equations of all straight lines in the current line channel;
the target tracking straight line selection module is used for selecting a target tracking straight line according to the straight line equations of all straight lines in the current line channel;
the deviation angle judging module is used for judging whether the deviation angle of the target tracking straight line is smaller than or equal to a preset standard deviation threshold value or not;
and the flight prediction module is used for predicting the flight prediction position of the aircraft by using the target tracking straight line if the deviation angle is judged to be less than or equal to the preset standard deviation threshold value.
10. A system for power line lockdown tracking processing using a lidar, comprising:
a memory, a processor and a program stored on the memory and executable on the processor for a power line lock tracking process using lidar, the program being executed by the processor to implement the steps of the method for a power line lock tracking process using lidar as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210475307.1A CN114859368A (en) | 2022-04-29 | 2022-04-29 | Method and system for tracking and processing power line locking by using laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210475307.1A CN114859368A (en) | 2022-04-29 | 2022-04-29 | Method and system for tracking and processing power line locking by using laser radar |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114859368A true CN114859368A (en) | 2022-08-05 |
Family
ID=82635937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210475307.1A Pending CN114859368A (en) | 2022-04-29 | 2022-04-29 | Method and system for tracking and processing power line locking by using laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114859368A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115760725A (en) * | 2022-11-04 | 2023-03-07 | 广东安恒电力科技有限公司 | Power transmission line external force invasion monitoring method, medium and equipment based on laser radar |
| 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 |
-
2022
- 2022-04-29 CN CN202210475307.1A patent/CN114859368A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115760725A (en) * | 2022-11-04 | 2023-03-07 | 广东安恒电力科技有限公司 | Power transmission line external force invasion monitoring method, medium and equipment based on laser radar |
| CN115760725B (en) * | 2022-11-04 | 2024-02-20 | 广东安恒电力科技有限公司 | Laser radar-based transmission line external force intrusion monitoring method, medium and equipment |
| 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 |
|---|---|---|
| CN106022381B (en) | Automatic extraction method of street lamp pole based on vehicle-mounted laser scanning point cloud | |
| CN108875666B (en) | Method and device for acquiring motion trail, computer equipment and storage medium | |
| US9142011B2 (en) | Shadow detection method and device | |
| CN114859368A (en) | Method and system for tracking and processing power line locking by using laser radar | |
| Mariescu-Istodor et al. | CellNet: Inferring road networks from GPS trajectories | |
| Verucchi et al. | Real-Time clustering and LiDAR-camera fusion on embedded platforms for self-driving cars | |
| CN111582054B (en) | Point cloud data processing method and device and obstacle detection method and device | |
| KR20140043393A (en) | Location-aided recognition | |
| WO2012155121A2 (en) | Systems and methods for estimating the geographic location at which image data was captured | |
| TWI581207B (en) | Computing method for ridesharing path, computing apparatus and recording medium using the same | |
| CN112883820A (en) | Road target 3D detection method and system based on laser radar point cloud | |
| CN112101374B (en) | Unmanned aerial vehicle obstacle detection method based on SURF feature detection and ISODATA clustering algorithm | |
| CN111783722B (en) | Lane line extraction method of laser point cloud and electronic equipment | |
| CN113901951A (en) | Unmanned ship self-identification and obstacle avoidance method and device | |
| CN114386466B (en) | Parallel hybrid clustering method for candidate signal mining in pulsar search | |
| CN113484875A (en) | Laser radar point cloud target hierarchical identification method based on mixed Gaussian ordering | |
| CN115115607A (en) | Image shape feature extraction and recognition method based on image analysis | |
| CN112287906B (en) | Template matching tracking method and system based on depth feature fusion | |
| CN114529883A (en) | Method and electronic device for detecting traffic light | |
| CN112907574B (en) | Landing point searching method, device and system of aircraft and storage medium | |
| Sayed et al. | Point clouds reduction model based on 3D feature extraction | |
| CN115187884A (en) | High-altitude parabolic identification method and device, electronic equipment and storage medium | |
| CN115994934B (en) | Data time alignment method and device and domain controller | |
| CN117864155A (en) | Laser radar-based road edge post-processing method and device, vehicle and storage medium | |
| CN113554072B (en) | Flight action dividing method, computer readable storage medium and terminal equipment |
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 |