Insulator Recognition Of The Aerial Power Line Image

Due to the electric power-line working for a long time in the bad natural environment, combined with long-term mechanical load and strong electric field and other factors, that makes the malfunction of the insulator, so we need checking on transmission lines regularly. But due to the laying of the transmission line passes through mountains and rivers and many other complex terrains, manual inspection is very difficult. At present, helicopter has been widely used for data collection in the inspection of the transmission line, and how to detect the fault of transmissions line from multifarious data is a focus in the study of the current difficulties.In this paper, the main work is to study the insulator’s segmentation and recognition of the aerial power line image, which also is the basis of fault identification. In image preprocessing part, considering the partial differential equation (PDE) on the edge of the denoising in keeping the good properties, as well as to avoid the stock effect using the fourth-order partial differential equations (Four-order PDE) method for the Gaussian noise filtering, and the median filtering method has been used to filter out impulse noise(salt and pepper noise). Finally, using the structural similarity index measurement (SSIM) evaluates the results.In the part of image segmentation, considering the shape characteristics of insulator, the method of the gray scale morphology combining the threshold is used to process the image.Through using several common segmentation methods at the same time, and comparing the results of the analysis, found that the most of the present methods can achieve good insulator image segmentation, though time-consuming is slightly longer than other methods, but still meet the needs of the project.Finally, in the section of the insulator identification, insulator identification method based on contour is proposed by studying the principle of generalized Hough transform. On the basis of good image segmentation, compared to the generalized Hough transform in improving the efficiency, this method can achieve a more accurate insulator positioning.