Obstacle detection using millimeter-wave radar and its visualization on image sequence

Sensor fusion of millimeter-wave radar and a camera is beneficial for advanced driver assistance functions such as obstacle avoidance and stop&go. However, millimeter-wave radar has low directional resolution, which engenders low measurement accuracy of object position and difficulty of calibration between radar and camera. In this paper, we first propose a calibration method between millimeter-wave radar and CCD camera using homography. The proposed method does not require the estimation of rotation and translation between them, or intrinsic parameters of the camera. Then, we propose an obstacle detection method, which consists of an occupancy-grid representation, and a segmentation technique, which divides data acquired by radar into clusters (obstacles); thereafter, we display them as an image sequence using calibration results. We demonstrate the validity of the proposed methods through experiments using sensors that are mounted on a vehicle.