Search Results

Driver state sensing technologies start to be widely used in vehicular systems developed from different manufacturers. To optimize the cost and minimize the intrusiveness towards driving, majority of these systems rely on in-cabin camera(s) and other optical sensors. With their great capabilities of detecting and intervening driver distraction and inattention, these technologies might become key components in future vehicle safety and control systems. However, currently there are no common standards available to compare the performance of these technologies, thus it is necessary to develop one standardized process for the evaluation purpose.

Vehicle safety systems may use occupant physiological information, e.g., occupant heights and weights to further enhance occupant safety. Determining occupant physiological information in a vehicle, however, is a challenging problem due to variations in pose, lighting conditions and background complexity. In this paper, a novel occupant height estimation approach is presented. Depth information from a depth camera, e.g., Microsoft Kinect is used. In this 3D approach, first, human body and frontal face views (restricted by the Pitch and Roll values in the pose estimation) based on RGB and depth information are detected. Next, the eye location (2D coordinates) is detected from frontal facial views by Haar-cascade detectors. The eye-location co-ordinates are then transferred into vehicle co-ordinates, and seated occupant eye height is estimated according to similar triangles and fields of view of Kinect.