Detection Method of Insulator String Pin Based on YOLOv5-nS Algorithm

Aiming at the difficulty and slow speed of multi-pose insulator pin detection in single-chip insulator replacement robot in substation insulator replacement robot, a fast and reliable insulator pin detection method YOLOv5-nS based on YOLOv5 is proposed. First of all, a new backbone network structure is proposed for the Backbone part of the original network, which is based on the improved nSNet of ShuffleNetV2; Secondly, the Neck of the network is lightweight while adding a feature fusion layer; Thirdly, the boundary box regression loss function is modified to improve the accuracy of the boundary box prediction; Finally, the Anchor scale is re-selected. The experimental results show that under the condition that the accuracy of the improved YOLOv5-nS algorithm model is basically unchanged, the parameters are reduced by 89.7%, the model size is reduced by 87.5%, and the frame rate is increased by 7%.