| With the development of Global Energy Internet technology and the upgrade of power grid itself,it is very important to ensure the security and stability of power system operation.Therefore,a higher standard is put forward for routine inspection and protection of transmission lines.Compared with the traditional manual inspection,UAV electric inspection is widely used in routine inspection because of its more advantages.This paper mainly studies the method of insulator location and fault detection based on multi-feature fusion.First of all,aiming at the problem of large noise and unclear of the aerial image sample,this paper adopts the corresponding image pre-processing,and designs a simple image preprocessing small program with the software,it provides a basis for insulator location and identification.Secondly,after comparing and analyzing the characteristics and effects of different image segmentation methods,Canny operator is selected to segment the edge of the image.In order to solve the problem of insulator identification in aerial images,an insulator location method based on morphological-spatial domain feature fusion is proposed.The image is transformed into the HSI color space which is more adaptive to human perception,and then the feature description parameters of the projection curve are calculated according to the regularity of the projection curve of insulator,the image region satisfying the defined discriminant is determined as the candidate region of the insulator.According to the characteristics of the ellipse shape of the insulator,an improved ellipse detection method based on Hough transform is used to determine the accurate region of the insulator.Then,based on the accurate region of insulator,a new method for detecting the dropout fault is proposed.After the image of insulator is processed by morphology,the fault area is judged by comparing the distance relationship between adjacent insulators.The fault recognition rate of this method can reach 92%,and it can complete the dropout fault detection accurately and efficiently.Finally,the interface of insulator location and fault detection is designed based on the above algorithm,which provides convenience for UAV electric power inspection. |
