A Multi-Target Detection Method on Distribution Cabinet Based on Improved Faster R-CNN

It is an inevitable trend to detect and recognize the states of different component on distribution cabinet panel more effectively accurately by using inspection robots instead manpower. Aiming at problems multiple recognition targets large size difference in image panel, improved Faster R-CNN multi-target detection method designed. Automatically required achieved this method. In paper, Resnet50 used VGG16 as feature extraction network R-CNN, Adam Optimizer Momentum, anchor box changed adapt target panel. Experiments show that model has higher accuracy less computational cost dataset.