Research Article
Application of Yolov5 Algorithm in Identification of Transmission Line Insulators
@INPROCEEDINGS{10.1007/978-3-030-89814-4_65, author={Jinxiong Zhao and Jiaxiu Ma and Junwei Xin and Rutai An}, title={Application of Yolov5 Algorithm in Identification of Transmission Line Insulators}, proceedings={Mobile Multimedia Communications. 14th EAI International Conference, Mobimedia 2021, Virtual Event, July 23-25, 2021, Proceedings}, proceedings_a={MOBIMEDIA}, year={2021}, month={11}, keywords={Transmission line Insulator Yolov5 Algorithm Data enhancement}, doi={10.1007/978-3-030-89814-4_65} }- Jinxiong Zhao
Jiaxiu Ma
Junwei Xin
Rutai An
Year: 2021
Application of Yolov5 Algorithm in Identification of Transmission Line Insulators
MOBIMEDIA
Springer
DOI: 10.1007/978-3-030-89814-4_65
Abstract
As an important infrastructure, the power system assumes a position that cannot be ignored in the national economy. The insulator in the transmission line is one of the main components of the power system. A complete and defect-free insulator is a prerequisite to ensure a good insulation between the current-carrying conductor and the ground. At present, it has become a mainstream practice to identify insulators through drones. However, due to the small number and single types of insulator data currently disclosed, the network does not have a large number of samples to learn more characteristics of insulators, which hinders the improvement of the accuracy of the network model to a certain extent. In this article, based on the existing 848 transmission line insulator data set, we train the yolov5 algorithm to generate a network with a recognition rate. The experimental results show that the mAP of the trained model is 11.41% higher than that of J-Method and 37.25% higher than the average of the other four methods mentioned by J-Method.
