A fault image recognition method for the breakage of the railway wagon hook lifter

Abstract

An image recognition method for a broken fault of a railway freight car hook lifter, which relates to the technical field of freight train detection, and aims at the problem that the traditional image processing detection method in the prior art is affected by mud and oil stains on the hook lifter itself, resulting in low detection efficiency. A method for image recognition of a broken fault image of a railway freight car hook lift rod, comprising the following steps: Step 1: Obtain an image of the area where the hook lift rod is located as a sample data set; Step 2: Carry out data amplification to the sample data set; Step 3: Analyze the data Mark the concentrated images; step 4: generate a data set from the original image and the labeled image, and train the model; step 5: use the UNet network model to predict the area of ​​the hook lift bar, and obtain the binary image of the hook lift bar; step 6: Determine whether the hook lifting rod is broken, if so, give a fault alarm, if not, go to step 5. The invention can improve the efficiency and accuracy of fault detection.

Description

technical field [0001] The invention relates to the technical field of freight train detection, in particular to an image recognition method for a broken fault image of a railway freight car hook and handle. Background technique [0002] The hook lifter is a part for uncoupling the couplers of two cars connected to each other. When freight cars are marshalling at the station, it is often necessary to turn the hook lifter to unmarshal the vehicles. If the hook lifter breaks, it will affect the train Marshalling operations. In the past, two detection methods, manual detection and traditional image processing methods, were used. The artificial detection method has the influence of human subjective factors, and the phenomenon of artificial fatigue will cause the occurrence of fault detection and missed detection. The traditional image processing detection method is aimed at the influence of mud and oil stains on the parts themselves, and the algorithm itself is not robust, whi...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to propose a fault image recognition method for railway freight car hook lift rod breakage in view of the problem that the traditional image processing detection method in the prior art is affected by mud and oil stains on the hook lift rod itself, resulting in low detection efficiency

Images