Panoramic Vision System to Eliminate Driver's Blind Spots using a Laser Sensor and Cameras

In this paper, we propose a multipurpose panoramic vision system for eliminating the blind spot and informing the driver of approaching vehicles using three cameras and a laser sensor. A wide-angle camera is attached to the car trunk and two cameras are attached under each side-view mirror to eliminate the blind spot of a vehicle. A laser sensor is attached on the rear-left of the vehicle to gather information from around the vehicle. The proposed system performs a pre-processing to estimate several system parameters. We compute a warping map to compensate the wide-angle lens distortion of the rear camera. We estimate the focus-of-contraction (FOC) in the rear image, and interactively compute the homography between the rear image and the laser sensor data. Homographies between each side-view image and the rear image are also computed in the pre-processing step. Using the system parameters obtained in the pre-processing step, the proposed system generates a panoramic mosaic view to eliminate the blind spot. First we obtain the undistorted rear image using the warping map. Then, we find road boundaries and classify approaching vehicles in laser sensor data, and overlap the laser sensor data on the rear image for further visualization. Next, the system performs the image registration process after segmentation of road and background regions based on road boundaries. Finally, it generates various views such as a cylindrical panorama view, a top view, a side view and an information panoramic mosaic view for displaying varied safety information.