CN114627374A - Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method - Google Patents
Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method Download PDFInfo
- Publication number
- CN114627374A CN114627374A CN202210234386.7A CN202210234386A CN114627374A CN 114627374 A CN114627374 A CN 114627374A CN 202210234386 A CN202210234386 A CN 202210234386A CN 114627374 A CN114627374 A CN 114627374A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- laser radar
- insulator
- point
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The application relates to a point cloud acquisition system and an insulator identification and positioning method based on a laser radar and a cloud platform, wherein the point cloud acquisition system comprises the laser radar, a computer and the cloud platform, and the laser radar is vertically arranged in the center of the cloud platform; the computer is respectively connected with the laser radar and the holder; the computer drives the laser radar to scan through the ROS, and simultaneously drives the holder to rotate at a constant speed at a fixed angle, so that the whole scene is scanned; after scanning, the laser radar transmits the point cloud data obtained by scanning to the computer through an Ethernet protocol for subsequent processing. The insulator identification method and the insulator identification device directly adopt the three-dimensional point cloud to identify the insulator, can directly output space coordinate information of the insulator while completing an identification task, and are high in processing speed and strong in pertinence.
Description
Technical Field
The application relates to the technical field of insulator identification and positioning, in particular to a point cloud acquisition system based on a laser radar and a cloud platform and an insulator identification and positioning method.
Background
The insulator plays a role of supporting a wire and insulation in the fields of a transformer substation, a power transmission line and the like, and thus is required to have good mechanical and electrical properties. Because the insulator bears the effects of cold and heat changes, mechanical stress, atmospheric pollution and the like for a long time in the operation process, the insulator needs to be overhauled and cleaned regularly to ensure the operation reliability. However, the manual live-line operation mode has high requirements on the technical level and skill level of operators, weather conditions, safety protection tools and the like, the robot live-line inspection and cleaning becomes the trend of power grid technology development, and the problem to be solved firstly is the identification and positioning problem of the insulator in the operation space.
The existing insulator identification method takes an image as a medium, and identifies insulator information in the image through an image processing algorithm, but the spatial position information of the insulator is difficult to be directly given, and other methods are needed to solve the spatial position information of the insulator so as to guide the motion of the robot, so that the calculation complexity is high. However, in other fields, a deep learning method is adopted for insulator identification and positioning, but the deep learning method requires a large amount of sample data for training, which is costly, and it is difficult to directly output spatial position information of an insulator.
Disclosure of Invention
An object of the embodiment of the application is to provide a point cloud collection system and an insulator identification and positioning method based on a laser radar and a cloud platform, wherein three-dimensional point cloud is directly adopted for identifying an insulator, the space coordinate information of the insulator can be directly output while the identification task is completed, the processing speed is higher, and the pertinence is stronger.
In order to achieve the above purpose, the present application provides the following technical solutions:
in a first aspect, an embodiment of the present application provides a point cloud collection system based on a laser radar and a pan-tilt, including a laser radar, a computer, and a pan-tilt,
the laser radar is vertically arranged at the right center of the holder;
the computer is respectively connected with the laser radar and the holder;
the computer drives the laser radar to scan through the ROS, and simultaneously drives the holder to rotate at a constant speed at a fixed angle, so that the whole scene is scanned; after scanning, the laser radar transmits the point cloud data obtained by scanning to the computer through an Ethernet protocol for subsequent processing.
The laser radar is a 360-degree laser radar, and the laser radar and the holder are both disc-shaped structures.
In a second aspect, an embodiment of the present application provides an insulator identification and positioning method based on a laser radar and a pan-tilt, including the following specific steps:
starting a laser radar and a holder, and scanning a large-range scene;
filtering the point cloud obtained by scanning based on prior knowledge to obtain an interesting area;
carrying out clustering analysis on the point cloud of the region of interest by using a region growing method to obtain different clustering clusters;
calculating the viewpoint histogram characteristics of different clustering clusters;
calculating Euclidean distances between the viewpoint histogram features of different clustering clusters and the insulator template;
and identifying and positioning the insulator through a given threshold value, and displaying in a scene obtained by scanning.
The specific method for filtering the point cloud obtained by scanning to obtain the region of interest comprises the steps of processing an original point cloud by a straight-through filtering method, establishing a channel based on a point cloud coordinate system by the straight-through filtering method, filtering the point cloud outside the channel and only reserving the point cloud in the channel, wherein the point cloud in the channel represents the region of interest.
The method comprises the steps of adopting a segmentation algorithm based on region growth to calculate curvature of all points in point cloud and sort the points, selecting the point with the minimum curvature as an initial seed point, designing an empty seed point sequence and an empty cluster array, setting the number of field points searched by the seed points, starting searching by the initial seed point, adding the field points into the current cluster array if the threshold value of the included angle between the normal line of the field points and the normal line of the current seed points is smaller than a given threshold value, checking the curvature of the current field points, adding the field points into the seed point sequence and deleting the current seed points if the curvature is smaller than the given threshold value, continuing to grow by new seed points until the seed point sequence is empty, completing the growth of a region, and finally repeating the operations on the rest points, until all point traversals are complete.
And identifying and positioning the insulators through a given threshold value, and displaying the insulators in a scene obtained by scanning, wherein the identification of the insulators is specifically to calculate the viewpoint histogram feature of each cluster point cloud by traversing the cluster points, calculate the Euclidean distance between the viewpoint histogram feature and the viewpoint histogram feature of the insulator template, and if the Euclidean distance is smaller than the given threshold value, consider the cluster point cloud as the insulator.
The positioning and displaying of the insulator in the scanned scene is specifically that the clustering point cloud identified as the insulator is enveloped by using an enclosure, the coordinate of the center point of the clustering point cloud under the point cloud coordinate system is calculated, and the clustering point cloud is displayed in the original point cloud, so that the positioning and displaying of the insulator are realized.
Compared with the prior art, the invention has the beneficial effects that: compared with the traditional insulator identification method based on image processing, the method has the advantages that the identification and the positioning adopt different media, the three-dimensional scanning of a large scene is carried out by adopting a laser radar and a cloud platform, the identification of the insulator is directly carried out by utilizing the three-dimensional laser point cloud obtained by scanning, and the spatial position information of the insulator can be directly output; compared with the identification method based on deep learning, the method does not need to spend a large amount of manpower and material resources to acquire and classify the data set, can realize the identification and the positioning of the insulator at a longer distance, and has higher identification precision.
Drawings
To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and that those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
FIG. 1 is a diagram of a point cloud acquisition system according to an embodiment of the present application;
FIG. 2 is a flow chart of the insulator identification and positioning method of the present application;
FIG. 3 is an original point cloud image of a large-scale scene reconstruction in an embodiment of the present application;
FIG. 4 is a point cloud image after x-axis straight-through filtering of an original point cloud;
FIG. 5 is a point cloud image after y-axis direct filtering of an original point cloud according to the present application;
FIG. 6 is a flow chart of the segmentation algorithm based on region growing according to the present application;
FIG. 7 is a flow chart of the insulator identification algorithm of the present application;
FIG. 8 is a front view of the positioning effect of the insulator under a wide range of scenes according to the present application;
FIG. 9 is an oblique view of the insulator positioning effect in a large-scale scene of the present application;
fig. 10 is the envelope box center point coordinates of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
In a first aspect, as shown in fig. 1, the invention provides a point cloud collection system based on a laser radar and a pan-tilt, comprising a laser radar 1, a computer 2 and a pan-tilt 3,
the laser radar 1 is vertically arranged at the right center of the holder 3;
the computer 2 is respectively connected with the laser radar 1 and the holder 3;
the computer 1 drives the laser radar 1 to scan through the ROS, and the computer 1 simultaneously drives the holder 3 to rotate at a constant speed with a fixed angle, so that the whole scene is scanned; after scanning, the laser radar 1 transmits the point cloud data obtained by scanning to the computer 2 through the ethernet protocol for subsequent processing.
Laser radar 1 is 360 degrees laser radar, laser radar 1 and cloud platform 3 are disc type structure.
As shown in fig. 2, in a second aspect, an embodiment of the present application provides an insulator identification and positioning method based on a laser radar and a pan-tilt, including the following specific steps:
starting a laser radar and a holder, and scanning a large-range scene;
filtering the point cloud obtained by scanning based on prior knowledge to obtain an interesting area;
carrying out clustering analysis on the point cloud of the region of interest by using a region growing method to obtain different clustering clusters;
calculating the viewpoint histogram characteristics of different clustering clusters;
calculating Euclidean distances between the viewpoint histogram features of different cluster clusters and the insulator template;
and identifying and positioning the insulator through a given threshold value, and displaying in a scene obtained by scanning.
As shown in fig. 3, 4 and 5, fig. 3 reconstructs original point clouds for a large-scale scene, and since the laser radar scans the whole three-dimensional scene, the number of the original point clouds is 962639, the point cloud information is redundant and contains a large number of regions which are not interested. Because the position distance between the laser radar and the insulator is relatively fixed, the original point cloud is processed by a direct filtering method. The straight-through filtering method is characterized in that a channel is established based on a point cloud coordinate system, point clouds outside the channel are filtered out, only the point clouds in the channel are reserved, the point clouds in the channel represent an area of interest, and the number of the point clouds is greatly reduced compared with that of original point clouds. According to the invention, the original point clouds are filtered on an x axis and a y axis, the channel range of the x axis is (3m, 7m), the point cloud image after the x axis direct filtering is shown in fig. 4, the point cloud number is 54766, the insulator is manually marked, the range of the y axis is (-2m, 2m), the point cloud image after the y axis direct filtering is shown in fig. 5, the point cloud number is 15390, and the insulator is manually marked.
Through filtering the original point cloud, the point cloud data is greatly reduced, but the point cloud data is still disordered, so that the filtered point cloud needs to be subjected to cluster analysis, and the point clouds of the insulators are classified into the same category. The method comprises the steps of adopting a segmentation algorithm based on region growing, calculating curvature of all points in point cloud, sequencing, selecting the point with the minimum curvature as an initial seed point, designing an empty seed point sequence and an empty clustering array, and setting the number of field points for seed point searching. And starting searching by using the initial seed point, adding the field point into the current clustering array if the threshold value of the included angle between the field point normal and the current seed point normal is smaller than a given threshold value, then checking the curvature of the current field point, adding the field point into the seed point sequence and deleting the current seed point if the curvature is smaller than the given threshold value, continuing to grow by using the new seed point until the seed point sequence is empty, and finishing the growth of one region. And finally, repeating the operation on the rest points until all the points are traversed, wherein a segmentation algorithm flow chart based on region growing is shown in fig. 6, the number of the seed point search fields is 30, the normal included angle threshold value is pi/60, and the curvature threshold value is 1.
The 21 point cloud cluster can be obtained through the steps, the view histogram feature of each cluster point cloud is calculated through traversing the cluster, Euclidean distance calculation is carried out on the view histogram feature of each cluster point cloud and the view histogram feature of the insulator template, and the view histogram feature is composed of a fast point feature histogram and view components. Assuming that each clustered point cloud is represented by a point set P, the s-th point of the point cloud is PsPoint p ofsNormal vector of (a) is nsPoint p ofsThe t-th proximity point of (a) is ptPoint p oftNormal vector of (a) is ntAt point psThe coordinate system uvw is defined above, and the base vector thereof is:
u＝ns
w＝u×v
based on the coordinate system uvw, point psAnd point ptNormal vector difference n ofjA set of feature point features SPFH ═ α, β, γ, d) can be described by the formula:
α＝cos-1(v·nj)
γ＝tan-1(w·nj,u·nj)
d＝‖ps-pt‖2
for point concentration point psIn other words, there are k adjacent points, and the fast point feature can be obtained by weighting the point feature of the point itself with the point features of the k adjacent points, and the calculation formula is:
in the formula, wiRepresents a point psThe distance from its i-th point of proximity, k, is 50.
For each component in the FPFH of each point in the point set P, dividing the component into 45 intervals, counting the occurrence times of each component in the corresponding 45 intervals, and then drawing a histogram of the components, so as to obtain a 180-dimensional fast point feature histogram vector.
The view component is then the view v through the set of calculated points PpWith each point p thereonsAnd dividing the included angle theta into 128 intervals, counting the occurrence frequency of each component in the 128 intervals, and drawing a histogram of each component to obtain a 128-dimensional viewpoint component histogram.
Combining the fast point histogram of 180 dimensions and the view component histogram of 128 dimensions results in a view histogram feature VFH of 308 dimensions.
Obtaining viewpoint histogram characteristic VFH of clustered point cloudiAnd viewpoint histogram feature VFH of template point cloudtemplateCalculating the Euclidean distance d between the two feature vectors:
if the Euclidean distance is smaller than a given threshold value, the clustering point cloud is considered as an insulator, and an algorithm flow chart of insulator identification is shown in FIG. 7, wherein the Euclidean distance threshold value is 100.
Enveloping the cluster point cloud identified as the insulator by using an enclosure box, calculating the coordinate of the central point of the cluster point cloud identified as the insulator in a point cloud coordinate system, and displaying the coordinate in the original point cloud, thereby realizing the positioning and displaying of the insulator, wherein the result is shown in fig. 8 and 9, the enclosure box represents the position of the positioned insulator, and three pairwise vertical straight lines represent the original point and the construction mode of the current point cloud coordinate system. The envelope box center point coordinates are shown in fig. 10.
According to the embodiment of the application, the point cloud acquisition system based on the laser radar and the cloud deck is designed, a hemispherical environment space can be effectively covered, large-range three-dimensional scanning of transformer substation and distribution network environments can be realized, point cloud acquisition of a plurality of insulators is finally realized, and the problem that the reconstruction range of the traditional laser radar is small is effectively solved;
the embodiment of the application provides a method for directly identifying and positioning an insulator by using three-dimensional point cloud, which can realize accurate identification of the insulator and output spatial position information of the insulator;
according to the embodiment of the application, the prior information of the distance is used for completing the rough positioning of the insulator, and compared with the integral positioning of other methods, the point cloud processing amount is less, and the efficiency is higher;
the embodiment of the application provides the method for completing the rapid matching and positioning of the insulator in the point cloud by using the three-dimensional point cloud template of the insulator, and compared with a method for training a model by using a data set, the method has stronger pertinence and stronger practical application feasibility;
according to the method, the space positioning of the insulator is completed by using the envelope box, compared with other methods, the method is easily affected by the sparse point cloud data by detecting the edge of the insulator, the method can completely extract the space position and the posture of the insulator, and the success rate is higher.
The key content of the invention is that the invention provides a novel point cloud acquisition and insulator identification and positioning method which is different from the traditional image-based insulator identification method, the three-dimensional point cloud is directly adopted for identifying the insulator, the space coordinate information of the insulator can be directly output while the identification task is finished, the processing speed is higher, and the pertinence is stronger.
The above description is only an example of the present application and is not intended to limit the scope of the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (7)
1. A point cloud collection system based on a laser radar and a cloud platform is characterized by comprising the laser radar (1), a computer (2) and the cloud platform (3),
the laser radar (1) is vertically arranged in the center of the holder (3);
the computer (2) is respectively connected with the laser radar (1) and the cloud deck (3);
the computer (1) drives the laser radar (1) to scan through the ROS, and the computer (1) simultaneously drives the holder (3) to rotate at a constant speed at a fixed angle, so that the whole scene is scanned; after scanning, the laser radar (1) transmits the point cloud data obtained by scanning to the computer (2) through an Ethernet protocol for subsequent processing.
2. The point cloud collection system based on laser radar and cloud platform according to claim 1, wherein the laser radar (1) is a 360-degree laser radar, and the laser radar (1) and the cloud platform (3) are both in a disc-shaped structure.
3. An insulator identification and positioning method based on a laser radar and a holder is characterized by comprising the following specific steps:
starting a laser radar and a holder, and scanning a large-range scene;
filtering the point cloud obtained by scanning based on prior knowledge to obtain an interesting area;
carrying out clustering analysis on the point cloud of the region of interest by using a region growing method to obtain different clustering clusters;
calculating the viewpoint histogram characteristics of different clustering clusters;
calculating Euclidean distances between the viewpoint histogram features of different clustering clusters and the insulator template;
and identifying and positioning the insulator through a given threshold value, and displaying in a scene obtained by scanning.
4. The method as claimed in claim 3, wherein the filtering is performed on the point cloud obtained by scanning to obtain the region of interest, and the specific method comprises processing the original point cloud by a direct filtering method, wherein the direct filtering method establishes a channel based on a point cloud coordinate system, and filters out the point cloud outside the channel while only retaining the point cloud inside the channel, and the point cloud inside the channel represents the region of interest.
5. The method for identifying and positioning insulators based on the laser radar and the pan-tilt according to claim 4, wherein the point cloud of the region of interest is subjected to cluster analysis by using a region growing method to obtain different cluster clusters, specifically, a segmentation algorithm based on the region growing is adopted, the curvatures of all points in the point cloud are calculated and sequenced, a point with the smallest curvature is selected as an initial seed point, then an empty seed point sequence and an empty cluster array are designed, the number of field points for searching the seed points is set, the initial seed point is used for searching, if the angle threshold between the normal of the field points and the normal of the current seed points is smaller than a given threshold, the field points are added into the current cluster array, then the curvature of the current field points is checked, if the curvature is smaller than the given threshold, the field points are added into the seed point sequence and the current seed points are deleted, and continuing to grow by using the new seed points until the seed point sequence is empty, completing the growth of one region, and finally repeating the operation on the rest points until all the points are traversed.
6. The insulator identification and positioning method based on the laser radar and the pan-tilt-zoom lens as claimed in claim 5, wherein the insulator identification and positioning are completed through a given threshold and displayed in a scanned scene, wherein the insulator identification is specifically that the view histogram feature of each clustered point cloud is calculated by traversing the clustered clusters, and the euclidean distance between the view histogram feature of each clustered point cloud and the view histogram feature of the insulator template is calculated, and if the euclidean distance is smaller than the given threshold, the clustered point cloud is considered as the insulator.
7. The insulator identification and positioning method based on the laser radar and the cloud deck as claimed in claim 6, wherein the positioning and displaying of the insulator in the scanned scene are specifically implemented by enveloping the clustered point cloud identified as the insulator with an enclosure, calculating the coordinates of the center point of the clustered point cloud under the point cloud coordinate system, and displaying the clustered point cloud in the original point cloud, so as to realize the positioning and displaying of the insulator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210234386.7A CN114627374A (en) | 2022-03-09 | 2022-03-09 | Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210234386.7A CN114627374A (en) | 2022-03-09 | 2022-03-09 | Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114627374A true CN114627374A (en) | 2022-06-14 |
Family
ID=81900190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210234386.7A Pending CN114627374A (en) | 2022-03-09 | 2022-03-09 | Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114627374A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116184357A (en) * | 2023-03-07 | 2023-05-30 | 之江实验室 | Ground point cloud data processing method and device, electronic device and storage medium |
| CN116246121A (en) * | 2023-05-12 | 2023-06-09 | 山东科技大学 | Point cloud data processing method based on improved furthest point sampling |
| CN116452604A (en) * | 2023-04-10 | 2023-07-18 | 南京邮电大学 | Complex substation scene segmentation method, device and storage medium |
-
2022
- 2022-03-09 CN CN202210234386.7A patent/CN114627374A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116184357A (en) * | 2023-03-07 | 2023-05-30 | 之江实验室 | Ground point cloud data processing method and device, electronic device and storage medium |
| CN116184357B (en) * | 2023-03-07 | 2023-08-15 | 之江实验室 | Ground point cloud data processing method and device, electronic device and storage medium |
| CN116452604A (en) * | 2023-04-10 | 2023-07-18 | 南京邮电大学 | Complex substation scene segmentation method, device and storage medium |
| CN116246121A (en) * | 2023-05-12 | 2023-06-09 | 山东科技大学 | Point cloud data processing method based on improved furthest point sampling |
| CN116246121B (en) * | 2023-05-12 | 2023-08-11 | 山东科技大学 | Point cloud data processing method based on improved furthest point sampling |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114627374A (en) | Point cloud collection system based on laser radar and cloud deck and insulator identification and positioning method | |
| CN107146277B (en) | Method for modeling transformer substation based on point cloud data | |
| CN106709946B (en) | LiDAR point cloud-based automatic multi-split conductor extraction and fine modeling method | |
| CN112381861B (en) | Forest land point cloud data registration and segmentation method based on foundation laser radar | |
| CN110288586A (en) | A kind of multiple dimensioned transmission line of electricity defect inspection method based on visible images data | |
| CN107045733B (en) | Method for modeling GIS (gas insulated switchgear) of transformer substation based on point cloud data | |
| CN112633382B (en) | Method and system for classifying few sample images based on mutual neighbor | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN111914813A (en) | Power transmission line inspection image naming method and system based on image classification | |
| Zhang et al. | Research on spatial positioning system of fruits to be picked in field based on binocular vision and SSD model | |
| Tu et al. | Extraction of street pole-like objects based on plane filtering from mobile LiDAR data | |
| CN109284409B (en) | Picture group geographical positioning method based on large-scale street view data | |
| CN113033386B (en) | High-resolution remote sensing image-based transmission line channel hidden danger identification method and system | |
| CN116452604B (en) | Complex substation scene segmentation method, device and storage medium | |
| CN113295142A (en) | Terrain scanning analysis method and device based on FARO scanner and point cloud | |
| CN115937492B (en) | Feature recognition-based infrared image recognition method for power transformation equipment | |
| CN117115390A (en) | Three-dimensional model layout method of power transformation equipment in transformer substation | |
| CN113723468B (en) | Object detection method of three-dimensional point cloud | |
| CN116704497A (en) | Rape phenotype parameter extraction method and system based on three-dimensional point cloud | |
| Yuan et al. | Automatic cables segmentation from a substation device based on 3D point cloud | |
| CN114494586B (en) | Lattice projection deep learning network broadleaf branch and leaf separation and skeleton reconstruction method | |
| Damodaran et al. | Extraction of Overhead Transmission Towers from UAV Images | |
| CN113326892A (en) | Relation network-based few-sample image classification method | |
| Yang et al. | A robust model fitting-based method for transmission line extraction from airborne LiDAR point cloud data | |
| CN111435537A (en) | Model training method and device and pose optimization method and device based on splicing map |
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 |