Recognition of insulator based on developed MPEG-7 texture feature

Abstract

The well working status of insulators guarantees the safe-running of power system. It is significant to development patrol power transmission line system with UAV, since intelligent supervisory of insulator is very important for power system in future. This technical note firstly presents the preprocessing of insulator images. Using the improved MPEG-7 EHD (edge histogram method) technique, the method of extracting edge feature of insulator is determined. The identification of insulator image with template matching method is proposed. Finally, the real-time object tracking algorithm based on Kalman filter is derived. The possible position of the insulator in the next frame is predicted by Kalman filter's predictive function, and the global searching of an image is converted to local searching. An illustrative simulation result is given to demonstrate the effectiveness and feasibility of the proposed tracking algorithm.