Open AccessResearch on Detection Method of Insulator Image Based on Improved Faster R-CNN
Author(s) -
Samphors Pha,
Hanbo Zheng,
Yonghui Sun
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2213/1/012036
Subject(s) - convolutional neural network , computer science , insulator (electricity) , artificial intelligence , image (mathematics) , pattern recognition (psychology) , physics , optoelectronics
This study proposes the use of Improved Faster Region-Convolutional Neural Network (R-CNN) in target detection for insulator images in power systems. Faster R-CNN is essentially a combination of the Fast R-CNN and the Regional Proposal Network (RPN). The Faster R-CNN method, which is being used today, is an extension and improvement of the Fast R-CNN. The Improved Faster R-CNN is particularly highly effective in the detection of occluded targets and those with different aspect ratios and scales. Through the various experiments conducted in the study, it has been demonstrated that the improved Faster R-CNN is highly effective in detecting insulator images with different scales and with different aspect ratios. Furthermore, the improved Faster R-CNN is also very effective in detecting mutually occluded insulator images.