Abstract
Road accidents cause a great loss to human lives and assets. Most of the accidents occur due to human errors, such as bad awareness, distraction, drowsiness, low training, and fatigue. Advanced driver assistance system (ADAS) can reduce the human errors by keeping an eye on the driving environment and warning a driver to the upcoming danger. However, these systems come only with modern luxury cars because of their high cost and complexity due to several sensors employed. Therefore, camera-based ADAS are becoming an option due to their lower cost, higher availability, numerous applications and ability to combine with other systems. Targeting at designing a camera-based ADAS, we have conducted an ethnographic study of drivers to know what information about the driving environment would be useful in preventing accidents. It turned out that information on speed, distance, relative position, direction, and size and type of the nearby objects would be useful and enough for implementing most of the ADAS functions. Several camera-based techniques are available for capturing the required information. We propose a novel design of an integrated camera-based ADAS that puts technologies-such as five ordinary CMOS image sensors, a digital image processor, and a thin display-into a smart system to offer a dozen advanced driver assistance functions. A basic prototype is also implemented using MATLAB. Our design and the prototype testify that all the required technologies are now available for implementing a full-fledged camera-based ADAS.