Research On Overhead Transmission Lines Abnormal Detection Algorithm Based On UAV Aerial Image

Transmission line is an important component part of power system,the quality of its operation and maintenance is directly related to the safe and stable operation of power system.Transmission lines suffer from lightning strikes,external damages and other damage of natural environment for a long time.The transmission lines are extremely prone to two kinds of common abnormal conditions: broken strands and foreign bodies hanging on it.If it is not found and eliminated in time,it will inevitably affect the safe and stable operation of the power grids.At present,the electric power department usually uses the UAV to inspect instead of manual patrol.The inspector judge abnormal conditions on the massive aerial transmission line images with eyes.The results is not only poor efficiency,but also high-probability misjudgement.Therefore,this paper focuses on how to use image processing technology to automatically detect abnormal conditions on the image of the transmission line,and designs an abnormality method for overhead transmission lines based on UAV aerial images.The main research contents of this paper are as follows:Aiming at the drawbacks of the traditional Canny edge detection algorithm in the edge detection of aerial transmission line images,such as the difficulties of extracting small edges and consuming time to set a fixed threshold,the edge detection algorithm based on Canny and Edge Drawing(hereinafter referred to as ED-Canny)is optimized and proposed in this paper.The hybrid filter composed of the median filter and the gaussian filter is designed to achieve image denoising,and according to the relationship between the sizes of filter template and the effects of smoothing image,and the gaussian convolution kernel size of the hybrid filter is modified in order to achieve the purpose of retaining more image details while denoising the image.The weight determination factor parameter is set to search the boundary of the aerial image to obtain anchor points.The intelligent routing algorithm is used to connect the pixels between the anchor points in order to achieve the purpose of extracting the edges of the continuous closed image as completely as possible.The effective extraction of the image edge is completed ultimately.Aiming at the problems such as wire breakage,discontinuity and misdetected line in the process of wire extraction of transmission lines under complex backgrounds,the traditional phase grouping algorithm is improved based on the analysis of the distribution characteristics of transmission lines.The best phase partition method is proposed and the false segment criterion is designed to improve the algorithm.Firstly,the best phase partition method is used to obtain line supporting area,group and classify all the pixels in the image.Then the least square method is used to extract the straight line segments.Secondly,the mechanism of false segment criterion is used to screen the fitted line segments.Qualified wire segments are refitted.The improved algorithm not only solves the problem of wire breakage and discontinuity in the process of extracting straight lines,but also avoids the problem of misdetected line in the process of extracting straight lines and optimizes the efficiency of algorithm execution.Aiming at the two kinds of common abnormal conditions: broken strands and foreign bodies hanging on the transmission lines,the fault recognition model is designed.Firstly,this paper designs the model of sliding pixel block,sets the parameter of gradient rotation angle,which locates the precise location of the two common abnormal conditions through changing the rotation gradient angle of the sliding pixel block.Secondly,the fault recognition model is designed to detect broken strands,and the relevant parameters for abnormal discrimination are set to detect the large target foreign bodies attached to the transmission line,and the projection method is used to detect the small target foreign bodies attached to the transmission line.Finally,the template matching method is used to reduce the interference of detection results caused by the false foreign bodies.The experimental results are provided to verify the algorithm designed in this paper.The results show that the algorithm can fast and accurately detect abnormal conditions on the transmission line.