Forward obstacle detection in a lane by stereo vision

We have developed the forward obstacle detection system by using stereo vision. This system consists of a combination of a stereo processing system based on a Field-Programmable Gate Array (FPGA) and Central Processing Unit (CPU) on PC. By using the FPGA-based system, the system can process images of high resolution and frame rate in real time. Therefore, large field of view and accuracy of distance go together. This system defines the drivable tunnel by detecting road surface and lane. Then, the forward obstacle in the tunnel can be detected. It was confirmed that this system works by the experiment in the expressways.