Rear object detection method based on optical flow and vehicle information for moving vehicle

Abstract

This paper proposes a method to detect rear object using vehicle information and optical flow in moving vehicles. The previous method detects objects by method using optical flow, background removal, and frame difference in the stopped vehicle state. In particular, optical flow method uses analysis of background or flow distribution to remove the background flow. In this paper, however, objects detect in moving vehicles by removing the background flow generated by vehicle movement using vehicle information. The performance of the proposed algorithm evaluated through experiments using stored images and vehicle information obtained from actual moving vehicles.