Design and implementation of an image-based backing assistance system

This paper proposes an image-based vehicle safety system, which not only provides images of backing car scenarios but also detects obstacles actively at the same time. In this research, image processing techniques are implemented to carry out the goal of obstacle detection. A novel image force field applied on Active Contour Model (ACM) is proposed to segment obstacles fast and efficiently. This system has been implemented on low-cost DSP platform and experiment result shows warning accuracy is 93.12%.