Hough Transform Tuned Bayesian Classifier for Overhead Power Line Inspection

Algorithms for automatic electrical power line detection are investigated. In particular, this paper presents a novel tuning technique using the Hough-Transform to improve Bayesian pixel classification accuracy. Connected component analysis is used to remove small components (misclassified pixels) from the classified images. Experimental results presented show that the proposed algorithm outperforms traditional Bayesian classification for power line detection. 1