Research On Safe Distance Detection Method Of Substation Equipment Based On Machine Vision

Substation equipment is generally charged,and the voltage is high,which is easy to cause discharge and electric shock accidents.Therefore,it is especially important to monitor the safety distance between people or objects and substation equipment in real time.This paper proposes a machine vision-based safety distance detection method for substation equipment,which realizes the safety distance detection between the target and the substation equipment through the binocular vision system,so as to discover the potential danger in time.In this paper,the binocular vision system model is established firstly,and the binocular camera is calibrated by the traditional calibration method,then the internal and external parameters of the binocular camera are obtained.Simultaneously we get the distortion correction and parallel correction.The image quality is improved by preprocessing such operations as filtering,deblurring and brightness equalization.This pre-processing operations improve the image quality;then the monitoring video is processed in real time,the improved background subtraction method is used to identify and extract the moving target,and the Gaussian mixture filtering shadow suppression method is adopted to improve the accuracy of moving target detection;The basic algorithm related to stereo matching is adopted,and the ORB stereo matching algorithm is used to perform stereo matching on the left and right image pairs of the binocular camera at the same time,which effectively reduces the calculation amount and improves the speed and accuracy of binocular stereo matching.Finally,the distance between the target and the substation equipment is calculated by using the parameter matrix of the camera inside and outside and the feature pair of stereo matching,and the real-time monitoring of the safety distance is realized.Experiments show the feasibility and effectiveness of the method.