Research On Infrared Image Segmentation For Substation

With the scale of China's electric Power network continues to expand,it's essential that the substations operate securely and reliably, and the operation of electrical equipments is one of the key factors that determine its stable and secure operation. The Infrared diagnosing techniques,which have been more and more popularly applied,played an important role in Fault diagnosing for electrical equipments .Because of the electrical equipment variety and each kind of equipment's fault characteristic also different, the research on image segmentation to the infrared image is very important to carry on the electrical equipment fault diagnosis.The paper focuses on analyzing the application of Infrared image segmentation technology in the substation. According to the features of substation infrared image, the paper gets the best pre-processing program based on experimental results by using a variety of de-noising. For electrical equipment infrared image, the paper proposes a method of edge detection and mathematical morphology combined segmentation method. The experiment results show that this method is fit for the application in substation. By taking the complexity and uncertainty of the infrared image, the genetic neural network is proposed to segment the infrared image. Optimization of weights and thresholds in neural network based on genetic algorithm is executed to improve the convergence speed of the BP neural network. Compared with the standard BP neural network, the segment speed of the genetic neural network adopted is much higher.Level set method describes the evolution of geometric active contour by a compact mode and provides a stable numerical algorithm.The energy function of Mumford-Shah model synthetically uses the information of image's boundary and region,and its contour evolution is not concerned to the gradient of boundary.The complex computing restricts the application of image segmentation method based on Chan-Vese(C-V) level set model. In order to improve the speed of image segmentation, this paper proposed the time-consuming re-initialization and the corresponding iterative termination criterion, which can improve the speed and reduce the number of iterations. Experiments show that the method is simple and efficient, with good separation effects.