车载红外图像的行人检测与跟踪技术

摘要

针对车载红外图像的特点,采用基于区域估计的帧差法实现车载红外图像中的行人检测,再通过行人的几何特征和运动速率特征进行行人的识别,最后运用轮廓矩匹配算法进行行人的跟踪.实验结果表明所采用的方法具有较好的行人检测及跟踪效果.%Based on the characteristics of vehicle infrared images we detect pedestrians by the vehicle infrared images using three-frame-difference method which is based on region estimation. Then we make pedestrian recognition based on the geometric features and motion characteristics. Finally,pedestrian tracking is implemented using contour moment matching method. The experimental results show that the proposed method can detect and track the pedestrians accurately.