Research On X Ray Inspection And Image Processing Technology For Basin Insulator

In recent years,X ray digital imaging(X-DR)technology has been applied to the GIS equipment detection in China.This technology provides a new detection means for the detection and maintenance personnel,and also provides a new method for the state monitoring and maintenance of the GIS equipment.Since the time of X-DR detection technology is not long to be introduced in China,there are few researches on the platform building of the insulator in the GIS device,the recommendation standard of the detection parameters,the evaluation system of the quality of the X ray image,the establishment of the defect image library,the image enhancement and the defect recognition algorithm.In order to better and faster apply the X-DR detection technology to the detection of the basin insulators,the main defects of the basin insulators are simulated on the basis of a large number of relevant literature and field simulation tests,and the recommended parameters for obtaining the X ray detection of the basin type insulators with different voltage levels and the improvement of the image quality are obtained.The method is researched to achieve rapid diagnosis.In this paper,a detection platform for basin insulators is built,including indoor and outdoor detection platform.The platform can well satisfy the test requirements for different specifications basin type insulators.According to the different indoor and outdoor conditions,different protective measures have been determined,and three kinds of fault models of basin insulator have been made.In this paper,orthogonal test is applied to obtain the recommended parameters of X-DR ray detection in basin insulators.This test scheme can reduce the number of tests to a large extent,and fully protect the health of the experimenters.At the same time,this paper also proposes a set of X-ray image quality evaluation system for the basin insulator detection.This system not only refers to the objective image quality but also fully considers the subjective evaluation,and can reflect the image quality more comprehensively.By using the orthogonal test data analysis software,the recommended parameters of X ray detection for basin insulators with different voltage levels can be obtained.In view of the low contrast degree of X-ray images of the basin insulators,the image enhancement of the basin insulators is studied by using the adaptive histogram equalization algorithm with limited contrast.The simulation results show that the algorithm can effectively improve the image quality of the basin insulators.How to establish a perfect basin type insulator defect image library through the limited basin insulator defect image,the mutual information antagonism generation network depth learning technology is used to simulate the generation of new samples to enrich the basin insulator defect image library.The simulation results show that the mutual information antagonism generation network technology can be found.It is well applied to the extension of the defect image library.The method of defect diagnosis for basin insulator is studied.The defect recognition methods combining SIFT with LSSVM and HOG with SVM are modeled and simulated.The results show that the combination of HOG and SVM can identify the defects of basin insulators efficiently and accurately.