Rail fastener abnormity detecting system based on point light source array linear array imaging

Abstract

The invention rail fastener abnormity detecting system based on point light source array linear array imaging. The rail fastener abnormity detecting system comprises a point light source array, a linear array camera, a wheel encoder, an RFID detector and an industrial personal computer, wherein the point light source array projects an I-shaped circular light spot to a fastener to modulate the shape and height information of the fastener, the imaging area of the linear array camera penetrates the projection area of the I-shaped circular light spot, the shape modulation light strip image on the surface of the fastener is obtained through scanning, and a deep learning classification method is used to detect abnormities such as fastener elastic rod losing, fastener elastic rod loosening and fastener gasket losing. The rail fastener abnormity detecting system is low in cost, applicable to fastener abnormity detection under a high-speed traveling state and capable of effectively detecting various abnormities of different types of fasteners.

Description

technical field [0001] The invention relates to a rail fastener abnormality detection device, in particular to a rail fastener abnormality detection system based on point light source array line array imaging. Background technique [0002] Fasteners are components that connect rails and sleepers to form rail rows, and play an important role in ensuring track stability and reliability. In terms of railway infrastructure testing, my country has long been dominated by manual and static testing, with high maintenance costs, high strength, and poor safety. In recent years, my country's high-speed railways have developed rapidly, and higher requirements have been put forward for the automation and real-time performance of railway testing. The existing fastener detection technology at home and abroad is: continuous scanning device based on linear array laser, such as German Sick company; computer vision detection device based on area array image sensor, such as VIS system of Americ...

AI Technical Summary

Problems solved by technology

The disadvantages of this type of method are: 1) When using an area array camera to shoot the fastener area, in order to ensure that the rail fasteners are not missed in the images captured by the area array camera, the imaging frequency of the camera should be the frequency of the fasteners appearing during the train running. 2 to 4 times. When the train is running at high speed, for example, when the speed is 300km / h, taking the fastener spacing as 0.5m as an example, and the frequency of fastener appearance is 166Hz, the imaging frame rate of the camera is 332-664fps, which gives the imaging system Image transmission, storage, and processing pose great challenges, and high-speed imaging equipment is expensive; 2) Due to the transmission deformation of area array cameras, in order to ensure that the shooting fasteners have the same image size, an accurate imaging trigger system is required to ensure that the captured images It is located directly above the fastener, which is difficult to achieve when the train is running at high speed; 3) The proposed image processing algorithm is only for the W-shaped spring ( figure 1 (a)) fasteners are valid, but not valid for other types of fasteners, for example: for DZⅢ type fasteners ( figure 1 (b)), the algorithms of the two patents cannot handle the absence of DZⅢ type fasteners ( figure 2 (a)), semi-loose ( figure 2 (b)) and missing gaskets ( figure 2 (c)) problem, so the image processing algorithms of patents 2011100073637.X and 201410667430.9 are not universal

Images