Point switch notch detection method based on CNN and image processing

Abstract

The invention provides a point switch notch detection method based on CNN and image processing, and belongs to the technical field of detection. The method comprises the following steps: acquiring a switch machine gap image of a train passing through a turnout and a switch machine driving turnout conversion process; establishingd and training a convolutional neural network CNN ; inputting the acquired notch images into a trained CNN network for notch image recognition and classification, and outputting categories of the notch images, the categories of the images including six categories: positioning and reverse notch images of ZD6, S700k and ZYJ7 point switches; according to the category of the notch image, processing the notch image through an image processing algorithm, extracting notchfeature straight lines l_1 and l_2, wherein the image processing algorithm comprises image graying, focus area segmentation, image enhancement, image binarization and image notch feature straight linefitting; and calculating a notch value w according to a linear equation of the notch characteristic straight line. Notch detection systems of various types of point switches can be integrated into one system, and the universality and transportability of the system are improved.

Description

technical field [0001] The invention relates to a switch machine gap detection method, in particular to a switch machine gap detection method based on CNN and image processing, and belongs to the detection technical field. Background technique [0002] A turnout is a device in a railway that enables a train to transfer from one track to another. The switch machine drives the point rail of the switch to move to the position or reverse position to connect different tracks. The degree of closeness between the tip rail and the basic rail is directly related to the safety of the train. If the gap is too large, there will be a phenomenon of "four openings", which may lead to a train derailment accident in severe cases. The gap is reflected by the gap inside the switch machine, and the real-time detection of the size of the gap is very important for the railway signal system and emergency response. [0003] The traditional switch machine gap detection method is manual inspection o...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a switch machine gap detection method based on CNN and image processing, to at least solve the problems in the prior art that the accuracy rate decreases and the false alarm rate increases after the system runs for a period of time, and is developed for different types of switch machines The problem of high system development and maintenance costs

Images