Insulator recognition based on mathematical morphology and Bayesian segmentation

In the inspection of high-voltage transmission line, in order to be able to accurately and complete identification of insulators, it proposes an insulator recognition method based on mathematical morphology and Bayesian segmentation. After the insulator image is preprocessing procedures of image gray and image noise reduction, firstly the bias segmentation algorithm is proposed based on mathematical morphology and Bayesian segmentation to segment the insulator image. Then, a combined morphological filtering method is adopted to obtain the image segmentation results for further eliminating the non-target image, which is the combination of the closed-open. Finally, the gray level information of the filtered image is mapped to the original image, and the closed operation of mathematical morphology is carried out for the restored image. The image is the result of the insulator recognition. Through the experiment proves that this method can accurately and effectively identify insulator in transmission line image.