移动机器人行人目标活动区域快速检测算法

摘要

In the pedestrian detection of the moving robot, tracking scheme will fail on account of the displacement of motion object and background. In this paper, a saliency optical-flow de-tection algorithm is proposed for detecting and demarcating the active area of the indoor pedestri-an targets. This algorithm can provide the initial localization information quickly and efficiently for the visual detection and tracking algorithm implemented in moving robot. The simulation ex-periments show that this algorithm can detect the active area of pedestrian target quickly under the complex condition indoors. And the algorithm can be used in character recognition and ob-stacle-avoiding so that it has strong engineering value.%针对移动机器人运动目标检测过程中运动目标与背景均存在位移致使跟踪失效的问题，提出一种显著光流分析法对室内行人目标活动区域进行检测与标识，能够快速有效地为移动机器人视觉检测与跟踪算法提供简便的初始区域定位信息。经仿真实验分析，该算法能够在室内复杂的情况下，快速检测出行人目标的活动区域，方便人物识别与躲避，具有较强的工程应用价值。