Identification detection method for high speed railway overhead line supporting device parts

Abstract

The invention discloses an identification detection method for high speed railway overhead line supporting device parts. The method comprises the following steps that: a training sample library of a high speed railway overhead line supporting device image is established, wherein the training sample library comprises the coordinate information of each part which is manually circumscribed in the image as a detection target, and a category; a deep convolutional neural network based on a Faster-RCNN (Region Convolutional Neural Network) algorithm is established; a training sample in the above training sample library is input into the established Faster-RCNN network to finish model training; and an image to be detected is input the trained model to obtain an identification detection result of the high speed railway overhead line supporting device parts. By use of the method, through the deep convolutional neural network of a candidate area, a target to be detected is subjected to feature learning and target classification, a huge workload for manually identifying the fault of the high speed railway overhead line supporting device parts is greatly reduced, the automatic analysis of a field image is realized, identification classification can be carried out for various overhead line suspension gear parts, and the method is high in identification accuracy.

Description

technical field [0001] The invention relates to the technical field of deep learning and automatic detection of electrified railways, in particular to a method for identifying and detecting components of a catenary support device for high-speed railways. Background technique [0002] The pantograph-catenary system is a key component of the electrified railway system, which undertakes the important work of transmitting the electric energy in the traction network to the electric locomotive. Due to the complex mechanical and electrical interaction between the pantograph and the catenary device, catenary failures account for a large proportion of various equipment failures in electrified railways, seriously affecting the safety of electrified railways. At the same time, catenary faults are widely distributed and difficult to detect. The traditional detection of catenary support and suspension devices relies on manual detection. Although these methods can guarantee a certain acc...

AI Technical Summary

Problems solved by technology

At present, the detection of some single parts has been realized, but the efficiency and accuracy of positioning and detection still have a lot of room for improvement

It is difficult to locate and detect multiple parts at the same time, especially for small-scale parts such as fasteners and connectors.

Images