Human position detection by capturing spectral contents of images representative of human skin

摘要

A camera arrangement (1) is mounted on a vehicle to detect a human (2) in images. The output of the camera provides a signal to a processor (4) which analyses the signal to identify any area or areas of the captured images which have a specific spectral content representative of human skin. This task may be done in H, S, V colour space (fig.. 2). The processor(4) then determines the position of such areas. The processor may be adapted to control or modify the actuation of one or more safety devices (5). It may also determine successive positions of such areas to determine movement of such areas. The camera may be directed towards a seat or a space in front of the vehicle and the safety device may be pedestrians protection device (35, fig. 10). The processor may use triangulation to determine the distance from the camera to the identified areas.