Rail turning identification method based on prediction weight

Abstract

The invention belongs to the technical field of image recognition, and relates to a rail turning recognition method based on prediction weight. The method comprises the following steps that a camera device captures a front target image, acquires an ROI area, performs histogram equalization, performs edge detection by adopting Sobel transformation based on a multi-angle template, performs binaryzation by adopting a Bernsen algorithm based on median and Gaussian filter linear superposition, detects a straight line by adopting Hough transformation based on prediction weight, and performs edge detection by adopting a Bernsen algorithm based on Gaussian filter linear superposition. And counting the proportion of the straight line passing through the central point of the track. Compared with theprior art, the method has the advantages that the detection rate of the track curve is high, and the method is suitable for multi-track scenes. The front track can be recognized more effectively, a driver is reminded, the driver can make a correct judgment, and better guarantee is provided for train operation.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and relates to a track turning recognition method based on prediction weight. Background technique [0002] In recent years, with the implementation of rail transit construction planning, the construction of rail transit and passenger dedicated lines is developing at a high speed. The requirements for rail vehicles and operating equipment are also getting higher and higher. During normal operation, the train operation automatic protection system (ATP) is installed on the train equipment for automatic safety protection. However, for railway maintenance engineering vehicles, considering the particularity and economy of operation, no automatic protection system device for train operation has been installed. Especially when the front rail turns, it is necessary to manually judge whether there is a turn ahead and slow down manually, which greatly reduces the efficiency. Moreover, people c...

AI Technical Summary

Problems solved by technology

"The patent application number is 201410546719, the patent name is: an image recognition method for the minimum curvature radius of the track, the method includes "obtaining the target image through the camera device; performing histogram equalization preprocessing on the target image; constructing the equalized image Regional growth model, determine the similarity criterion, and determine the growth stop condition; perform curve fitting on the rail model to obtain the rail recognition image; judge whether there is a turn ahead." Both of the above two patents have carried out preprocessing and curve fitting on the rail, But there are some problems: the first preprocessing method is relatively simple, and the recognition effect is not good; the effect of the second curve fitting is actually not very good in practice, and there are many interference noises, including factors such as multi-track railways and trees; the third In real-time photography, the information between two adjacent frames of images is very similar, and many existing methods ignore the feature point information of two adjacent frames of images

Images