Rotation invariant power line insulator detection using local directional pattern and support vector machine

Detecting and localizing insulator plays a vital role in any power line monitoring system. In this work, we present a novel method for rotation invariant insulator detection. Rotation invariance is achieved by an efficient approach for estimating rotation angle of all insulator of an image. Sliding window based local directional pattern (LDP) feature is extracted from the image and support vector machine is used for classing each of those sliding window. We demonstrate our approach on an evaluation set of 325 real-world insulator images captured from a moving vehicle and evaluate our results with respect to a manually created ground-truth. The performance of our insulator detector is comparable to other state-of-the-art detectors.