An active contour approach to water droplets segmentation from insulators

Abstract

Faulty insulators may compromise the electrical and mechanical integrity as well as safety of a power delivery system leading to leakage currents or causing voltage drop in the power grid. It is therefore important to monitor and inspect the insulator to assess damage that could be caused by aging or any accident on power line system. Computer vision has been identified and is being investigated as a tool to solve this problem safely, accurately and speedily. Hydrophobicity of insulator has been identified as an important means to ascertain the status of insulator. In this paper an active contour approach is used in water droplets segmentation on insulator. Experiment shows an improved performance when morphological operation is applied with the active contour model. Results achieved are evaluated qualitatively and quantitatively using a manually created ground-truth. Results obtained are presented and discussed.