Transmission line extraction and recognition from natural complex background

With increasing of the use of helicopters for transmission line inspection, the image processing technology is used to inspect the status of the transmission lines. In this paper, an algorithm of transmission line automatic extraction and recognition from natural complex background in aerial images is presented. A developed Marr-Hildreth edge detecting algorithm was used to enhance the linear features of the original image, and Hough transform was used to extract the line and curve from the image, at last, the morphological analysis is used to recognize the transmission lines between the transmission lines and background objects. The experimental results show that the proposed method can successfully extract and recognize the transmission lines regardless of background complexity, and it also proves that the algorithm is effective and practical.