Recognition And Location Of The Insulators Fault Based On Binocular Vision

The insulator is an important part of the transmission line in the safe operation.The safety accidents due to the failure of the insulator account for a large proportion of the transmission line faults.Through processing and analyzing the inspection image of the insulator,it is possible to detect the operation status of the insulator in time and diagnose the type and level of the fault in real time.This paper presents an identification and localization algorithm of the insulator fault based on binocular vision,which can improve the accuracy and real-time performance in recognition and detection of the insulator fault.In this paper,a binocular vision system model was established.Binocular cameras were calibrated based on the calibration method of Zhang Zhengyou,and the images were corrected by the principle of parallel calibration.According to binocular image features,the image preprocessing method was studied.Through filtering,brightness equalization,sharpening and other operations,the quality was improved and the features were highlighted of images.Combining with OTSU,morphology and minimal external moment method,the insulator was identified firstly,and then based on the ORB stereo matching algorithm,the feature of binocular image was extracted and matched which reduced the computation effectively and improved the speed and accuracy of matching binocular image.Finally,common faults of insulators were identified and positioned,the three-dimensional information of faults were measured according to the principle of binocular vision,the types and degrees of faults were analyzed quantitatively and measured accurately through the complementary features of binocular visual information.The experimental results showed that compared with monocular images,the recognition correctness and measurement of binocular image accuracy are higher,and the false recognition rate was reduced significantly.It is of practical significance and feasibility in the qualitative and quantitative analysis of insulator faults.