Nighttime Pedestrian Ranging Algorithm Based on Monocular Vision

Since the traditional computer vision ranging algorithm is imperfect in pertinence and precision, night time monocular vision pedestrian ranging method is proposed for vehicular infrared night vision goggles. Firstly, the method calibrated the internal and external parameters of infrared night-vision goggles, then, it corrected distortion of collected Vehicular Infrared Night Vision Image, and finally it ranged objective pedestrians by using night time monocular vision pedestrian ranging algorithm. The experimental results show that this method has the characteristics of pertinence, high precision and good real-time, and has good practicability.