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

Power transmission line hidden danger detection method based on instance segmentation

A technology of transmission lines and detection methods, applied in image analysis, image enhancement, instruments, etc., can solve the problems of incompatibility between processing accuracy and efficiency, and achieve the effect of improving operating efficiency and accuracy

Active Publication Date: 2020-06-26
SHANDONG UNIV +3
View PDF14 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The invention solves the technical problem in the prior art that processing accuracy and efficiency cannot be taken into account when performing target detection and recognition on UAV images

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 transmission line hidden danger detection method based on instance segmentation
  • Power transmission line hidden danger detection method based on instance segmentation
  • Power transmission line hidden danger detection method based on instance segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0074] Such as figure 1 , 2 , 3 shown. A method for detecting hidden dangers of transmission lines based on instance segmentation, comprising the following steps:

[0075] S1: Use data enhancement technology to expand the self-constructed anti-vibration hammer data set, and preprocess the anti-vibration hammer image data according to the visual characteristics of anti-vibration hammer corrosion;

[0076] S2: Build a feature extraction network:

[0077] Including improving the feature extraction network of ResNeXt+ Feature Pyramid Network (Feature Pyramid Network, FPN) according to the characteristics of the shock hammer target;

[0078] The network parameter value trained on the COCO dataset is used as the initial parameter of the network through the method of transfer learning, and at the same time, the structure of the feature extraction network is adjusted by reducing the convolutional network level to improve the operating efficiency of the algorithm;

[0079] S3: Modi...

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 power transmission line hidden danger detection method based on instance segmentation, which uses an improved MASK R-CNN model to perform automatic detection of power transmission line hidden dangers in the power field based on an instance segmentation technology, and especially has obvious advantages in automatic detection of stockbridge damper corrosion. For an image shot by an unmanned aerial vehicle, a target area is acquired through the model, and accurate target feature extraction and classification detection can be realized. Based on a target instance obtainedby segmentation, an image processing technology is used to detect a target fault hidden danger degree, and target detection result weighting calculation is combined so that a target fault hidden danger level can be accurately quantified. And meanwhile, the characteristics of the detection target can be considered. According to the method, priori knowledge is utilized to perform image processing, feature extraction network structure adjustment and non-maximum suppression algorithm improvement, and a method of combining instance segmentation and target detection is introduced, so that automaticdetection of the target is realized, and the accuracy of the hidden danger target in the power transmission line is improved.

Description

technical field [0001] The invention discloses a method for detecting hidden dangers of transmission lines based on instance segmentation, and belongs to the technical field of intelligent detection of transmission lines. Background technique [0002] High-voltage overhead transmission lines are exposed to complex natural environments, and are affected by wind, sun, rain, etc. for a long time, and are prone to various hidden dangers. For example, the surface of related hardware such as anti-vibration hammers matching lines is prone to rust. Corrosion will affect the overall performance of the fittings and greatly reduce the connection strength between the fittings and other components. Corroded fittings are prone to displacement and deformation when encountering severe weather such as strong winds, and may even break and fall. This will not only affect the normal power transmission of the transmission line, but also pose a threat to the safety of pedestrians under the tower....

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
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06T5/40G06T7/11
CPCG06T7/11G06T5/40G06V20/13G06V2201/07G06N3/045G06F18/214Y04S10/50
Inventor 聂礼强郑晓云战新刚姚一杨吴建龙徐万龙
Owner SHANDONG UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com