Open AccessRailway obstacle detection based on radar and image data fusion
Author(s) -
Shuai Qi,
Yu Dong
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1965/1/012141
Subject(s) - obstacle , computer vision , artificial intelligence , computer science , radar , frame (networking) , object detection , sensor fusion , image (mathematics) , pattern recognition (psychology) , geography , telecommunications , archaeology
In order to improve the poor adaptability of a single sensor in detection and further improve the train’s ability to perceive the running environment ahead, a railway obstacle detection method based on the fusion of microwave radar and image data was proposed. The environment in front of the train collected by the radar is preprocessed to eliminate false and out-of-bounds target signals, and the target within the safety limit is retained. At the same time, the frame difference method is used to detect the obstacles in the image sequence captured by the vehicle camera. The detection results of radar and camera are fused at the decision level. The experimental results show that this method can effectively realize the joint detection of obstacles by microwave radar and machine vision, which makes up for the deficiency of single sensor in obstacle detection.