A UAV-Based Transmission Line Defect Detection Method

Aiming at the problems such as few defect types and no unified detection parameters, a transmission line defect detection method based on deep learning is proposed in this paper. Firstly, the power transmission line defect identification data set is established. Secondly, in order to further reduce the impact of background and lighting factors on the identification accuracy, a relevant data set is defined, and an end-to-end deep learning algorithm with high identification accuracy based on region-based convolutional neural networks (R-CNN) is proposed for establishing a detection model. Finally, experimental results show that the proposed detection method has a better detection performance.