摘要
混合交通是我国城市交通的主要特点。我国城市步行交通在总出行量中所占比例较大,且将持续相当长的一段时间。而随着智能交通系统(ITS)研究与应用的不断深入,在优化交通管理、减少交通事故、保证行人安全等方面的要求越来越高,所以行人识别具有重要意义。视频检测是一种在智能交通监控中广泛使用的检测方式。文中基于视频图像,利用行人的形状、速度、非刚体性的特点,设计了一种具有较好实时性和鲁棒性的行人识别算法。
Mixed traffic is a major characteristic of Chinese urban traffic. The pedestrian traffic accounts for a large proportion of it and will last quite a long time. With the research and application of Intelligent Transportation Systems （ITS） go deeper, the requirement of optimization of traffic management, reduction of traffic accidents and protection of pedestrian safety becomes higher, so the method of pedestrian recognition is of great importance. Video detection is a kind of intelligent traffic detection methods in widely use. Making use of the characteristics of shape, velocity and nonrigid characteristics of Pedestrian, a real-time and robust pedestrian recognition algorithm based on video is designed to achieve practical goals.
出处
《信息化研究》
2010年第3期19-21,共3页
INFORMATIZATION RESEARCH
关键词
行人识别
形状
速度
非刚体
pedestrian recognition
shape
velocity
nonrigid body
