A cascaded classifier-based detection method for the hexagonal nut falling off bad state of the double-sleeve connector of the high-speed railway catenary

Abstract

The invention discloses a cascaded classifier-based detection method for the defective state of a double-sleeve connector part of a high-speed railway catenary. It includes the following steps: first, establish a sample library about the double-sleeve connector parts, extract the HOG features of the samples to train the cascaded AdaBoost classifier, and train the support vector machine classifier on this basis to realize the positioning of the double-sleeve connector parts ;Secondly, the Hough transform is used to extract the inclination angles of the flat arm and oblique arm connected by double sleeves in the target image, and the flat arm is rotated to the horizontal direction, thereby realizing the segmentation of the flat arm and the connector; and then the oblique The arm rotates to the vertical direction to realize the division of the inclined arm and the connector. When determining the fault, in the extracted connector image, set the horizontal baseline according to the position of the light spot, detect the singular value of the horizontal gray scale, and make an accurate judgment on the failure of the hexagonal large nut of the double sleeve connector.

Description

technical field [0001] The invention relates to the field of fault detection of high-speed railway catenary, in particular to a method for detecting the bad state of hexagonal nut shedding of the high-speed railway catenary double-sleeve connector based on a cascade classifier Background technique [0002] In the L-shaped arm support device of the high-speed railway catenary, the double-sleeve connector is an important load-bearing component. In order to ensure the safety of the train, the construction quality of this component has strict requirements. For dual ferrule connectors, the hex nut is an important fastener. The vibration or construction defects generated during the long-term operation of the train may cause the large nut of the casing to fall off, which reduces the load-bearing capacity of the wrist arm, reduces the mechanical strength of the catenary, and increases the possibility of accidents. The 4C system technical specification promulgated by the former Mini...

AI Technical Summary

Problems solved by technology

Since the images of catenary support and suspension devices collected on site are generally complex, and the fasteners in the photos are easily affected by reflections, it is difficult to use image processing technology to detect them. At present, there are no relevant reports on this aspect of research.

Images