基于特征的地铁门与安全门间隙异物检测方法

A Foreign Object Detection Way Based on Feature for the Gap between Subway and Safety Doors

在城市轨道交通运营过程中,存在人或物等异物被夹挤/遗留在安全门与地铁车门之间的可能性,造成安全事故.为此,提出一种基于关键特征匹配的地铁车门与安全门间隙的异物检测方法.该方法首先对检测图像与参考图像进行SURF特征提取与匹配后,引入一种基于局部信息的误匹配点剔除法,以提炼出用于后续异物检测的关键匹配点;然后,基于特征匹配点相对数量和特征点分布面积信息对待检测区域进行异物检测.试验结果表明,本文方法可有效地对地铁车门与安全门之间的间隙进行异物入侵检测,具有一定的实际应用价值.

During operation of urban rail transit,there is a possibility that people or objects will be squeezed/left between a safety gate and a subway gate.In this work,a foreign object detection method is proposed based on SURF feature extraction and matching algorithm for the gap between subway doors and safety doors.After the feature extraction and matching of the image to be detected and the reference image,a method of removing the mismatching points based on local information is given to extract the key points for subsequent foreign object detection.Then,a foreign object detection method in the area to be detected based on the relative number of matching points and the distribution area of feature points is proposed.The experiments show that the method can detect the foreign object intrusion of the foreign object detection method between subway doors and safety doors,and the proposed method is applicable.