StereoBox: a robust and efficient solution for automotive short-range obstacle detection
EURASIP Journal on Embedded Systems
GOLD: A Framework for Developing Intelligent-Vehicle Vision Applications
IEEE Intelligent Systems
Hi-index | 0.00 |
This paper presents a vision-based method able to increase safety in access automation systems. These systems include automatic swing and sliding gates, bollards and barriers to prevent unwanted access. In current installations the anti-crushing protection is ensured by an electronic device installed on the control boards, which directly controls drive torque, and a couple of infrared photodetectors: when an obstacle is touched by the gate leafs or barriers, or cuts the infrared beam, the control board stops the gate movement. Conversely, the new method proposed in this paper avoids collisions with obstacles and increases the overall detection performance. The new device returns a stop signal when an obstacle is present in a predefined area. The algorithm has been integrated in a real access system to test its performance.