Recognition of insulator explosion based on deep learning

Insulator is an extremely important component of the power transmission system. This article adopts the model of convolutional neural network from recent studies on deep learning to achieve end-to-end intelligent detection of insulators, which helps computers to identify the insulator from the foota...

Full description

Saved in:
Bibliographic Details
Published in2017 14th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) pp. 79 - 82
Main Authors Gao, Feng, Wang, Jiao, Kong, Zhizhan, Wu, Jingfeng, Feng, Nanzhan, Wang, Sen, Hu, Panfeng, Li, Zhizhong, Huang, Hao, Li, Jianqing
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2017
Subjects
Deep learning
Drones
Explosions
Fault detection
Feature extraction
Insulator
Insulators
Machine learning
Object detection
Semantic segmentation
Online AccessGet full text

Cover

More Information
Summary:Insulator is an extremely important component of the power transmission system. This article adopts the model of convolutional neural network from recent studies on deep learning to achieve end-to-end intelligent detection of insulators, which helps computers to identify the insulator from the footage faster and obtain fault detection more accurate of the insulator. Firstly, the method of object detection is used to determine the location of the insulator; then the insulator is extrapolated using fully convolutional networks; lastly, based on insulator's fault explosion characteristic, the coordinates of the fault explosion can be detected. The experimental results indicate that the method can effectively detect the faulted insulators in highly cluttered images, and our insulator fault detection outperforms existing methods.
DOI:10.1109/ICCWAMTIP.2017.8301453