Detection Of Icing-thickness On Transmission Line Based On CB Mathematical Morphology

Due to China’s vast, complex terrain, icing phenomenon on transmission line is more widely in our country, many areas in China, which have both icing disasters and accidents. Icing and snowing disasters can bring serious damage to national economy; affect the normal production and life of the people. Therefore, mastering the icing situation of transmission line at any time, real-time detection of the icing thickness, when the icing volume exceeds the preset amount, timely early warning signal, rapid de-icing measures plays a key role in disaster prevention and mitigation.Through the analysis of characteristics and geometric model of the transmission line image before and after the icing, a new edge detection icing-thickness method based on multi-scale Contour-Based mathematical morphology (short for CB morphology) is proposed in this paper. This method regroups and realizes a modified morphological algorithm, through full using the priorities of several traditional morphology algorithms. A new edge detector is defined by using the multi-scale operation of the modified morphological algorithm (MMN operator), and the edge contours of the images before and after icing are compared to calculate the actual icing thickness.In this paper, the main work is as follows:(1) In-depth study the formation mechanism of icing on transmission line and insulator, and the research status of the detection method of icing on insulator and transmission line and the image detection method at home and abroad.(2) Designing of on-line detection device system, studding the characteristics of image of transmission insulator and lines and putting forward the geometry model of icing thickness.(3) According to the characteristics of icing image of transmission line and insulator before and after, multi-scale on-off filtering algorithm and MMN edge detection operator is proposed. Through gray scale, the multi-scale on-off filtering and improved multi-scale CB morphology edge detection method to extract the edge image of before and after icing, and based on the geometric model of (2) to calculate the thickness of the ice.(4) Through comparing the results of artificial measuring and the algorithm, and comparing the effect of before and after the improved algorithm, verifying the accuracy and effectiveness of the proposed algorithm.