A Vision Based Top-View Transformation Model for a Vehicle Parking Assistant

Abstract

This paper proposes the Top-View Transformation Model for image coordinate transformation, which involves transforming a perspective projection image into its corresponding bird's eye vision. A fitting parameters searching algorithm estimates the parameters that are used to transform the coordinates from the source image. Using this approach, it is not necessary to provide any interior and exterior orientation parameters of the camera. The designed car parking assistant system can be installed at the rear end of the car, providing the driver with a clearer image of the area behind the car. The processing time can be reduced by storing and using the transformation matrix estimated from the first image frame for a sequence of video images. The transformation matrix can be stored as the Matrix Mapping Table, and loaded into the embedded platform to perform the transformation. Experimental results show that the proposed approaches can provide a clearer and more accurate bird's eye view to the vehicle driver.