Estimation of high voltage insulator contamination using a combined image processing and artificial neural networks

In this paper, contamination level estimation tool for high voltage insulators has been developed. A digital camera has been used to capture pictures. Image processing has been used to extract needed features form the captured images. Two types of features were considered. The first is “histogram based statistical feature” while the second is “singular value decomposition theorem based linear algebraic feature”. Using extracted features, a neural network has been successfully designed to correlate the insulator captured image and the contamination level. Testing of the developed estimation tool showed a very high successful rate in estimating the contamination level of unseen insulators. It is expected that a successful deployment of the developed tool will eliminate the need of human intervention in determining the time and location of insulators to be washed.