Power transmission tower bird nest fault detection method based on YOLO

Abstract

The invention discloses a power transmission tower bird nest fault detection method based on YOLO, and belongs to the technical field of computer vision. The detection method comprises the following steps: data acquisition: enabling an unmanned aerial vehicle to photograph an image of a power tower and covers different angles; marking the collected data set, and randomly dividing the data set intoa training set, a verification set and a test set according to the ratio of 8: 2: 1; clustering the marked target boxes through a K-means algorithm to obtain improved anchor points; training the training set and the verification set through a YOLO neural network to obtain a final model of the YOLO neural network; and detecting the test set by using the YOLO neural network detection model obtainedin the step 4. Compared with a conventional YOLO classification network, the method is more outstanding in recognition efficiency and accuracy. The detection speed is high, and the application requirement of power line unmanned aerial vehicle normalized inspection can be met; the detection accuracy is high, the robustness is high, and the speed is high.

Description

technical field [0001] The invention belongs to the technical field of computer vision, in particular to a bird's nest fault detection method based on a yolo neural network. Background technique [0002] As of the end of 2018, the loop length of transmission lines above 35kV across the country has reached 1.892 million kilometers. While birds bring peace and joy to humans, they also bring many troubles and hazards to the power system. According to operating experience and statistical data, the frequency of power failures caused by bird activities is second only to lightning strikes and external damage. Bird damage specifically includes: bird body short circuit failure, bird nesting, bird predation, guano flashover, bird natural enemy body short circuit, bird pecking composite insulator failure, etc. Iron towers have a much higher failure rate of bird damage than electric poles, and 90% of the failures occur in iron towers. The reason is that iron towers are tall and stable,...

AI Technical Summary

Problems solved by technology

In recent years, power grid companies around the world have increasingly applied unmanned aerial vehicles (UAVs) to line inspections. However, the amount of data collected by drones is large, and relying solely on manual identification is inefficient and prone to errors. Therefore, it is necessary to use an intelligent target detection algorithm

Images