Power transmission line hardware corrosion detection and analysis method based on deep learning

Abstract

The invention discloses a power transmission line hardware corrosion detection and analysis method based on deep learning, and particularly relates to the technical field of image recognition, and the method comprises the steps: 1, information collection: installing a high-definition camera on an unmanned aerial vehicle, carrying out the high-definition photographing of a power transmission line hardware through the operation of the unmanned aerial vehicle; and step 2, information analysis: identifying and analyzing the picture obtained in the step 1 by using a computer; and step 3, monitoring and alarming. According to the method, the faster-rcnn is introduced to position the position of the hardware fitting, the retrieval range is narrowed, multiple detections are greatly reduced, and comprehensive judgment is carried out by combining whether multiple feature target areas are rusted or not, so that false detections are greatly reduced, the detection probability of the hidden danger of corrosion of the power transmission line hardware fitting is improved, and support is provided for avoiding power transmission line safety incidents caused by corrosion of the hardware fitting.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a deep learning-based method for detecting and analyzing corrosion of metal fittings in transmission lines. Background technique [0002] Target detection is one of the basic tasks in the field of computer vision, and the academic circle has nearly 20 years of research history; in recent years, with the rapid development of deep learning technology, target detection algorithms have also shifted from traditional algorithms based on manual features to based on Detection technology of deep neural network; [0003] With the continuous development of target detection technology, its application fields are also expanding, and the detection of transmission lines is also applied to target detection technology; when detecting transmission lines, the main detection is the power on the transmission line Fittings, the power fittings are metal accessories made of iron, aluminum or ...

AI Technical Summary

Problems solved by technology

[0006] For this reason, the present invention provides a method for detecting and analyzing the corrosion of metal fittings of transmission lines based on deep learning. By introducing faster-rcnn to locate the position of the fittings, the search range is narrowed, and multiple inspections are greatly reduced, and whether the target area is rusted or not is combined with various features. Carry out comprehensive judgment, thereby greatly reducing false detections, improving the detection probability of transmission line hardware corrosion hidden dangers, providing support for avoiding transmission line safety incidents caused by hardware corrosion, and solving some deep learning network operations existing in the existing technology The speed is slower, the selection of kernel function is more difficult than the method of linear support vector machine, the method based on statistics will cut the picture into many small pieces, and the calculation speed will also be reduced, and the method based on directionality may be difficult to adapt to various shapes of fittings And the problem of insufficient characteristics considered;

Images