Aiming at the gap between subway screen door and train door, there is a hidden danger of accident in subway operation because of the pinch-in, and a detection method based on metro station surface is proposed. This method improves the efficiency and accuracy of the detection by employing a manual construction of the optical background and using the HSV color space. The algorithm uses the method to extract the target image of the car taillight with the camera of the vehicle head, and determines whether there is spatial foreign body in the gap between the car body and the screen door by calculating the completeness of the target. Experiments on real video data show that the proposed algorithm has good robustness to light changes and can detect vari-ous foreign objects accurately and can assist the driver in making decisions before driving.%针对地铁屏蔽门和列车门之间的间隙会产生夹人夹物等地铁运营安全事故隐患,提出一种基于K-means的地铁站台异物面检测方法.该方法通过采用人工构造光学背景,并且采用HSV颜色空间来提高检测的效率和准确度.算法采用K-means方法对车首摄像头拍摄车尾灯带图像进行目标提取,通过对目标的完整性计算来判断列车车体与屏蔽门缝隙是否存在空间异物.通过对真实视频数据进行实验,结果表明所提算法对光照变化具有很好的鲁棒性,可以准确检测出各种异物,能够辅助司机进行开车前的决策.
展开▼
