Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Laser point cloud ground object automatic identification method fusing line channel orthoimage

An orthophoto and automatic recognition technology, applied in character and pattern recognition, knowledge-based computer systems, machine learning, etc. Point cloud classification efficiency and other issues

Pending Publication Date: 2022-08-02
EXAMING & EXPERIMENTAL CENT OF ULTRAHIGH VOLTAGE POWER TRANSMISSION COMPANY CHINA SOUTHEN POWER GRID
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] (1) Although traditional algorithms such as unsupervised classification and supervised classification have great advantages in data processing efficiency, they are far from sufficient in classification accuracy, especially on orthophotos with centimeter-level accuracy. satisfactory
Although algorithms such as deep learning and machine learning have advantages in classification accuracy, the efficiency of data classification processing is not high.
[0012] (2) From the current process of image classification, it can be found that most studies focus on the classification and processing of data, while ignoring the quality of the data itself, and often the quality of data directly affects the results of image classification
[0013] At present, with the continuous improvement of UAV lidar technology, its application fields are also expanding, and it has good applications in power inspection, terrain mapping, 3D modeling and other fields. The point cloud data is unordered point cloud data, and the relevant information of the ground or specific objects cannot be obtained directly from the data. Data processing is required to extract effective information from the point cloud data.
The current point cloud data automatic recognition technology has certain shortcomings in the selection of point cloud classification accuracy and point cloud classification efficiency. If the accuracy of point cloud classification is to be improved, the efficiency of point cloud classification will be reduced. Efficiency will lose the accuracy of point cloud classification to a certain extent

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Laser point cloud ground object automatic identification method fusing line channel orthoimage
  • Laser point cloud ground object automatic identification method fusing line channel orthoimage
  • Laser point cloud ground object automatic identification method fusing line channel orthoimage

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0075] S1: Affected by weather, time, UAV flight conditions, etc., there will be color differences within a single image of UAV orthophoto data or between multiple images, as well as shadows of ground objects or orthophotos. Unnecessary shading that affects the classification, for the display of the towers and wires in the orthophoto of the transmission line channel on the image, the classification of the image will produce a huge image. For this reason, it is necessary to perform uniform light and color before classification. Correction of shadows and masked image preprocessing; including:

[0076] S11: Orthophoto uniform light and color, the process is as follows figure 2 shown;

[0077] In the process of orthophoto uniform light and color, an adaptive global-to-local color balance method is chosen to eliminate the influence of color differences between adjacent optical images. This method combines global and local optimization strategies. In the case of specifying a refe...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a line channel orthoimage fused laser point cloud ground object automatic identification method comprising the following steps: collecting line channel orthoimage data, and pre-processing the line channel orthoimage data; adopting a random forest algorithm to identify and classify the corrected orthoimage; the method comprises the following steps: collecting line channel laser point cloud data and tower standing book data, preprocessing the laser point cloud data, extracting ground point cloud and tower and power line point cloud, and taking the remaining laser point cloud data as target point cloud; and carrying out point cloud segmentation on a target point cloud by taking an orthoimage classification result as a constraint condition, and carrying out fine classification on a coarse classification result by adopting a random forest algorithm to obtain a point cloud recognition result. On the basis of various characteristics of the orthoimage, the two-dimensional advantage of the orthoimage and the three-dimensional advantage of the laser point cloud are combined, the point cloud classification precision is improved, point cloud classification parallel processing is added on the basis of orthoimage classification, and the point cloud classification processing efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of ultra-ultra-high voltage transmission lines, and more particularly to a method for automatic identification of laser point clouds and ground objects that integrates orthophoto images of line channels. Background technique [0002] According to the classification strategy of the classification method, the existing point cloud classification methods can be divided into two categories: hierarchical classification methods and simultaneous classification methods. The hierarchical classification method consists of two steps, filtering processing and object classification. In the filtering stage, the hierarchical classification method uses point cloud filtering algorithm to divide point cloud data into ground points and non-ground points. Commonly used point cloud filtering algorithms include mathematical morphology algorithm, slope filtering algorithm, and irregular triangulation progressive encryption. Algori...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/764G06V10/80G06V10/762G06V10/774G06V10/82G06K9/62G06N5/00G06N20/00G06V10/54G06V10/56
CPCG06V10/764G06V10/80G06V10/762G06V10/774G06V10/82G06N20/00G06V10/54G06V10/56G06N5/01G06F18/2415
Inventor 张兴华姜诚黄和燕王黎伟张福罗望春王鸿涛梁晖明莫兵兵李翔刘洪驿石志彬
Owner EXAMING & EXPERIMENTAL CENT OF ULTRAHIGH VOLTAGE POWER TRANSMISSION COMPANY CHINA SOUTHEN POWER GRID
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products