摘要： 介绍了一个使用三维激光摄像机对道路上的自行车辆和行人进行检测和跟踪的实时系统，系统主要分为物体识别模块和目标匹配跟踪模块两部分，前者采用了迭代自组织的数据分析算法（ISODATA算法）和多阈值分割方法，后者使用了一种新颖的将轨迹连贯性函数和卡尔曼滤波器相结合的多目标匹配跟踪算法。户外实验表明系统具有较高的识别率。

关键词: 运动人体, 实时跟踪, 卡尔曼滤波, 轨迹连贯性, 智能交通

Abstract: This paper presents a real-time system to detect and track bicyclists or pedestrians using a laser sensor. The system configuration consists of an object detection module and a tracking module. To detect moving bodies, iterative self-organizing data analysis techniques algorithm (ISODATA) and multi-level thresholding method are used in the detection module. To determining the trajectories of moving objects, a novel method, which is a combination of path coherence function and Kalman filter, is adopted in the tracking module for multiple objects matching and tracking. Based on outdoor scenes, experimental results demonstrate the system has high recognition rate in real-time performance.

中图分类号: 

• TP274.4