Power Line Detection and Tracking Using Hough Transform and Particle Filter

In this paper, a new method, based on particle filter approach, for detection and tracking of power lines using aerial video streams is presented. In the proposed algorithms, the gradient of image is computed using Sobel operator. Then an adaptive thresholding method is used to convert images to black and white to decrease the computation time. For detection and tracking of power lines, particle filter is used. The particles are defined in the Hough Space (ρ, θ). The particles' weights are determined using the number of detected points on the corresponding line in the image. The proposed algorithm has been successfully tested offline on 100 frames of images taken from a power line using a UAV.