Insulator detection method, device and equipment based on deep convolutional neural network

Abstract

The invention relates to an insulator detection method, device and equipment based on a deep convolutional neural network. The insulator detection method based on the deep convolutional neural networkcomprises the steps: collecting a plurality of original images, carrying out preprocessing on the original images, and constructing a sample training set; constructing a multi-angle candidate regionnetwork structure based on an RPN network, an inclination NMS algorithm and an angle factor; based on the multi-angle candidate region network structure, training the deep convolutional neural networkby using the sample training set to obtain a training model; and recognizing the to-be-recognized image by using the training model, and outputting an insulator image with a target frame if an insulator exists in the to-be-recognized image. Thus, the extraction precision of the insulator information is effectively improved, the situation that missing detection occurs to insulators with complex backgrounds and small targets during detection is avoided, and the situation that the detection effect is poor due to partial overlapping or shielding in insulator images is also avoided.

Description

technical field [0001] The present application relates to the technical field of insulator detection, in particular to an insulator detection method, device and equipment based on a deep convolutional neural network. Background technique [0002] With the continuous development of the country's economic level and science and technology, power consumption continues to rise. As the infrastructure of power transportation, the safety and stability of transmission lines is the basic guarantee for national power consumption. Insulators are important components of transmission lines, and they are responsible for electrical insulation isolation and mechanical protection and support. Since most insulators are exposed to the natural environment all the year round, they must withstand lightning, wind and snow, high temperature, etc. while bearing the original load. In extreme weather, insulators are prone to aging and damage, which brings great safety hazards to power transportation. ...

AI Technical Summary

Problems solved by technology

These factors make the detection efficiency of manual naked eyes low, the workload is huge, and it is easy to cause a large number of false detections and missed detections due to visual fatigue.

[0004] In related technologies, in the final result of detecting insulators based on convolutional neural network, the target frame is usually horizontal, and the problem of insulator tilting at various shooting angles is not considered, which easily leads to inaccurate information extraction of insulators , and, for insulators with complex shooting backgrounds and small targets, it is easy to cause poor detection results due to missed detection or partial overlap and occlusion

Images