基于改进ViBe的室内目标检测方法  

Indoor Pedestrian Detection Method Based on Improved ViBe

作　　者：熊玲 陈勇[1] XIONG Ling;CHEN Yong(College of Computer and Information Science,Chongqing Normal University,Chongqing 401331)

机构地区：[1]重庆师范大学计算机与信息科学学院,重庆401331

出　　处：《现代计算机（中旬刊）》2018年第11期31-34,共4页Modern Computer

摘　　要：对于室内动态环境中行人的准确检测一直是一个比较重要的难点,由于室内存在光照变化、阴影的出现以及各种遮挡物等复杂干扰。提出一种基于改进的视觉背景抽取法(ViBe)的室内运动目标检测方法,该方法分别从选择领域和残影的消除来改进,一是在背景模型初始化时选择的领域以及在图像边缘时选择边缘像素扩展的方式来提高背景模型的准确度;二是设置像素的滞留值为检测的背景像素和前景像素分别检测以更快地消除鬼影或残影。实验结果表明,该方法可以提高背景模型的精确度,也能有效地降低残影或鬼影的干扰,在室内场景中进行行人检测具有较高的准确性、鲁棒性。The accurate detection of pedestrian in indoor dynamic environment is always an important difficulty because of the complex interference such as the variation of illumination,the appearance of shadow and all kinds of obstructions.Proposes a method of indoor moving target de?tection based on improved visual background extraction method ViBe.The first one is to improve the accuracy of background model by se?lecting the field of background model initialization and the method of edge pixel expansion when image edge is selected.The second is to set the residual value of pixels as background pixels and foreground pixels respectively to eliminate ghost or shadow more quickly.The ex?perimental results show that this method can improve the accuracy of background model and reduce the interference of residual or ghost im?ages effectively.The pedestrian detection in indoor scene has high accuracy and robustness.

关 键 词：ViBe 残影 室内复杂环境 行人检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]