知識庫

Real-Time Obstacle Detection Method for urban traffic situations based on stereo vision (基於立體視覺的快速市區障礙物偵測方法)
  • 發布年度:2011
  • 主要類別:駕駛輔助
  • 次要類別:論文
  •  

    Many obstacle detection methods have been proposed and most of them apply multiple sensors to detect obstacles. In this research, we employ a stereo vision system composed of dual cameras to detect and estimate the distance of the obstacle. Binocular cameras are mounted on a specially designed mechanism to satisfy the geometric restrictions of the ideal stereo vision system. With the current design and when the distortion of images due to camera lens is corrected by calibration, the disparity image can be estimated by the pattern matching method and shown in pseudo-color. To detect and locate the obstacles, we propose a method capable of finding the obstacle from the range image. The main idea is to use a top-view, blob segmentation and pre-defined three-dimensional constraints to detect the obstacles. The stereo vision system was also tested under real traffic situations. The obstacle detection rate was about 90% at a nearly real-time image processing rate of 10 frames per second.
     
    Keywords: stereo vision; obstacle detection

© 2016 ARTC All Right Recerved. Design by Creatop.