Information collection unmanned aerial vehicle for steel rail surface defect detection based on deep learning network

Abstract

The invention discloses an information collection unmanned aerial vehicle for steel rail surface defect detection based on a deep learning network, and belongs to the technical field of steel rail surface detection.The information collection unmanned aerial vehicle comprises an electric tractor, an information collection trailer, an image information collection mechanism, a data transmission mechanism and a cloud server, the electric tractor drags the information collection trailer to move along the steel rail to be detected, workers are assisted in overhauling the steel rail, manual patrol is not needed, the working intensity of steel rail detection is greatly reduced, the electric tractor and the information collection trailer are of a separated structure, vibration of a driving mechanism of the electric tractor can be isolated, and the detection efficiency is improved. The image information collection mechanism can collect data more stably, the electric tractor and the information collection trailer can be replaced independently after being damaged, maintenance of equipment is facilitated, and the utilization rate of the equipment is increased.

Description

technical field [0001] The invention relates to the technical field of rail surface detection, in particular to an information collection drone for rail surface defect detection based on a deep learning network. Background technique [0002] During the operation of the train, defects such as scars, cracks, surface scratches, peeling and wrinkles will be produced on the surface of the rail, which may become a hidden danger to the safety of railway transportation. At present, the probability of the occurrence of internal defects in the rail has been greatly reduced, but the situation of rail fracture caused by surface defects of the rail is becoming more and more common. Therefore, defect detection of railway tracks is an important means to ensure the safety of railway transportation. [0003] The detection of track surface defects has always relied on manual inspection, which is inefficient, and the detection results are affected by the experience, responsibility, weather co...

AI Technical Summary

Problems solved by technology

The eddy current method is not affected by impurities such as oil on the surface of the rail. The probe does not need to touch the specimen, but it is easily affected by the fluctuation of the sensor. It is required that the distance between the sensor and the rail surface should not exceed 2mm

[0007] Both of the above two detection methods require inspectors to operate the equipment on site, which is time-consuming and laborious, which greatly reduces the detection efficiency of rail surface defects and increases the detection cost. For this reason, we propose a rail surface defect detection based on deep learning network Gathering drones with information

Images