Detection of Insulators on Power Transmission Line Based on an Improved Faster Region-Convolutional Neural Network

Detecting insulators on a power transmission line is of great importance for the safe operation systems. Aiming at problem missed detection and misjudgment original feature extraction network VGG16 faster region-convolutional neural (R-CNN) in face different sizes, order to improve accuracy insulators’ lines, an improved R-CNN algorithm proposed. The replaces backbone with Resnet50 deeper layers more complex structure, adding efficient channel attention module based mechanism. Experimental results show that performance has been effectively through improvement network. model trained training set consisting 6174 insulator pictures, tested testing 686 pictures. Compared traditional R-CNN, mean average precision increases 89.37%, 1.63%.