关键词: 复杂背景, 无人机, 航拍图像, 电力线, EDPF, Radon变换

Abstract: How to completely and accurately identify the power line from the aerial image of the drone with complex background has become one of the key issues for the unmanned inspection of the power line. After analyzing the characteristics of the power line in the aerial image, a new algorithm of power line detection and identification in complex backgrounds is proposed. Firstly, the original image is histogram equalized to improve the contrast ratio. Then, the edge image is detected by the EDPF algorithm improved by the LoG operator, with the background noise filtered out and the power line edge detected. Finally, the power lines in the image are completely extracted using the Radon transform and prior knowledge. The experimental results show that the proposed method has higher recognition accuracy, more complete recognition effect and better robustness compared with the traditional Canny operator and Hough transform and the LSD algorithm.

Key words: complex background, unmanned aerial vehicle, aerial image, power line, EDPF, Radon transformation