Recognition of Aerial Insulator Image Based on Structural Model and the Optimal Entropy Threshold Segmentation

Abstract

Insulator image recognition is an important research on the after-treatment of power grid patrol by UAV (Unmanned Aerial Vehicle). According to the characteristics of aerial insulator image, this paper centers on a new method of aerial insulator recognition based on the structural model and the optimal entropy threshold (OET) segmentation. Firstly, on the condition of the lack of insulator sample library, SketchUp software is used to build the structural model of the insulator and generate insulator simulation images. Those insulator simulation images are used to build the training sample set. Then the insulator image is grayed and enhanced. Next, OET Algorithm is applied to segment the insulator from backgrounds, and the mathematical morphology is applied to improve the segmentation results. Finally, the moment invariants of the insulator and background are calculated to train classifiers based on Adaboost Algorithm, and then a strong classifier is created. The example and testing results show that this method can effectively achieve insulator image recognition under complex backgrounds.