Track circuit box fault detection method, system and equipment and readable medium

Abstract

The invention provides a track circuit box fault detection method, system and equipment, and a readable medium. The method comprises the following steps: a to-be-processed image is obtained, the imageis input into a preset convolutional neural network model, and whether there is a fault track circuit box in the image or not is detected; if there is a fault track circuit box in the image, the convolutional neural network model outputs fault information of the fault track circuit box; and the image and the fault information are sent to a data storage module for maintainers to browse and check.Through track circuit box fault detection based on deep learning, mileage positioning image acquisition of a track circuit box can be replaced, and missing detection can be avoided. Meanwhile, based on the deep learning method, fault classification and recognition can be carried out on the collected image, the workload of manual image detection is reduced, the detection period of the track circuitbox is shortened, the fault accuracy of the equipment is improved, and the maintenance efficiency of the track circuit box is improved.

Description

technical field [0001] The present application relates to the technical field of image processing, and in particular to a method, system, device and readable medium for detecting a fault in a track circuit box. Background technique [0002] As the modern railway network becomes more and more developed, rail transit is an indispensable way of travel in people's work and life, and its safety issues have been paid more and more attention. As a component of the track circuit, the track circuit box is an important equipment for train operation safety. Faced with the increasing mileage of the railway network, the number of track circuit boxes is increasing day by day, and they are distributed along the railway line. It is particularly important for railway maintenance, especially for track circuit boxes. [0003] In the existing technology, the method of positioning and acquisition is mainly used to collect images containing track circuit boxes, and then the maintenance personnel ...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, the method of positioning and acquisition is mainly used to collect images containing the track circuit boxes, and then the maintenance personnel judge whether the track circuit boxes are faulty, but because the layout of the track circuit boxes is complicated and disorderly, there is no unified specification , so it is easy to miss the detection object

Images