Power line inspection image dense matching method

A dense matching, power line technology, applied in neural learning methods, inspection time patrols, instruments, etc., can solve the problem of not taking into account the importance relationship and weighting of different features, and achieve the effect of fast and efficient inspection work.

Pending Publication Date: 2021-02-26
GUANGDONG POWER GRID CORP ZHAOQING POWER SUPPLY BUREAU
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the PSMNet network does not take into account the importance relationship between different features, and does not weight the important features.

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
  • Power line inspection image dense matching method
  • Power line inspection image dense matching method
  • Power line inspection image dense matching method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] A method for intensive matching of power line inspection images is based on a cascaded multi-scale residual learning network. It adopts the cascade learning strategy and is divided into two networks for learning; the first network performs preliminary feature extraction on the input stereo image. and preliminary disparity estimation; the second network performs disparity refinement on the preliminary disparity estimation process on the basis of the first network to obtain the final matching result.

[0030] like figure 1 As shown, in the stage of preliminary feature extraction, a multi-scale convolutional neural network is used to obtain feature information of different scales, thereby increasing the extraction of different feature information by the network.

[0031] like figure 1 As shown, the scales used are 3x3, 5x5 and 7x7 respectively, so as to ensure the adaptability of the network to multi-scale features.

[0032] like image 3 As shown, in the first network,...

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 relates to the technical field of electric power inspection, in particular to a power line inspection image dense matching method, which adopts a cascade learning strategy and is dividedinto two networks for learning. A first network performs preliminary feature extraction and preliminary parallax estimation on the input three-dimensional image; and a second network performs parallax refinement on the preliminary parallax estimation process on the basis of the first network, thereby obtaining a final matching result. According to the power line inspection image dense matching method, automatic inspection work of power line cruising can be achieved, large-scale inspection work can be rapidly and efficiently achieved, and under the effect of multi-scale convolution, the matching result can adapt to matching changes of different scales.

Description

technical field [0001] The invention relates to the technical field of electric power inspection, in particular to a method for intensive matching of images of power line inspection. Background technique [0002] Compared with manual line inspection, drone imagery for power line inspection has many advantages such as high efficiency and safety. Among them, dense matching using aerial images to obtain pixel-by-pixel disparity values ​​is an important step in restoring the true 3D model along the power line. In the process of realizing the dense matching of UAV images, it can be divided into dense matching based on traditional algorithm and dense matching based on deep learning algorithm according to whether the deep learning method is adopted. [0003] Traditional image dense matching methods often suffer from a series of ill-posed problems such as uneven illumination, low texture or repeated texture, which leads to matching accuracy and robustness. In recent years, with th...

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): G06K9/62G06N3/04G06N3/08G07C1/20
CPCG06N3/08G07C1/20G06N3/045G06F18/22
Inventor 陈亮夏国飞董承熙谭麒李俊宏曾彦超李广俊
Owner GUANGDONG POWER GRID CORP ZHAOQING POWER SUPPLY BUREAU
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap