Research On Transmission Line Anomaly Detection Algorithm Based On Digital Image Processing

As a transmission channel between the power plant and consumers,the transmission line has an irreplaceable role in the transmission of electrical energy.However,due to the impact of extermal environmental factors,such as lightning strikes,bird drop,wind and rain erosion,abnormal conditions such as strand breaking of the transmission line and attachment of foreign objects have occurred from time to time.This seriously threatens the stable operation of transmission lines.Based on the above reasons,this paper uses a camera fixed on the transmission tower to collect images and transmit them to the server through 4G wireless signals.On the basis of analyzing the characteristics of the collected transmission line images,a transmission line anomaly detection algorithm based on digital image processing technology is designed to realize online detection of transmission line anomalies.The online detection system improves the efficiency of transmission line inspection and ensures the efficient operation of the transmission grid.The research content can be divided into the following four aspects:(1)In the pre-processing part of the transmission line image,the characteristics of the collected transmission line image are analyzed,and the effects of the soble operator,Laplace opeator,LOG operator,Canny operator,Robert operator,etc.applied to image processing are different.The results show that Canny edge detection results are the most delicate,and can detect small edges.(2)In the extraction part of power transmission lines,the methods of Radon transform,Hough transform,and LSD transform used for wire extraction are first introduced.Aiming at the problem of unsatisfactory extraction of transmission lines with more complex background,a detection method based on improved Canny edge detection and Beamlet transform is designed.Aiming at more false edges and misdetected wires,the wire fitting and deduplication algorithms are designed.(3)For the identification of foreign objects in power transmission lines,the vector field based on the gradient of the image block is defined based on the idea of vector fields such as electromagnetic fields.The offset angle is used to locate the foreign objects or broken stock positions and make detection.At the same time,a detection method based on local projection method is designed for small targets such as bird drop and glitch damage.(4)Conduct a cross point detection for a very important interference model,namely the wire cross model.Pulling wires privately,restrictions on shooting angles,broken strands,etc.can all cause crossovers in the captured images.Cross-point detection can avoid being treated as a stock break,so as to avoid false positives.The anomaly recognition algorithm proposed in this paper can accurately realize the anomaly recognition of transmission lines,and provide a new solution for the online detection of the transmission line operation status.Therefore,the research in this paper provides a theoretical basis and implementation for the transmission line intelligent online detection system,further improves the power system intelligent detection system,and promotes the development of grid inspection and detection in the direction of intelligence.