面向智能监控的运动目标轮廓可靠提取  

A Reliable Contour Extraction Method for Motion Object in Surveillance Systems

作　　者：夏海英[1,2] 颜远辉 黄思奇 肖雯静[1] 

机构地区：[1]广西师范大学电子工程学院,广西桂林541004 [2]广西师范大学药用资源化学与药物分子工程教育部重点实验室,广西桂林541004

出　　处：《广西师范大学学报（自然科学版）》2014年第4期39-44,共6页Journal of Guangxi Normal University:Natural Science Edition

基　　金：国家自然科学基金资助项目(21327007);广西自动检测技术与仪器重点实验室(YQ14202);药用资源化学与药物分子工程教育部重点实验室资助课题(CMEMR2014-B15);广西高等学校科研资助项目;广西师范大学博士启动基金资助项目

摘　　要：针对目标运动检测算法得到的运动目标轮廓存在轮廓不完整、不准确等问题,本文提出一种基于Vibe算法和GVF snake的运动目标轮廓可靠提取算法。首先通过Vibe算法快速定位运动目标的感兴趣区域,并结合数学形态学、边界快速粗定位算法,获得运动目标轮廓粗定位,再利用改进主动轮廓模型能够实时可靠提取运动目标的轮廓。改进的提取算法具有良好的模型初始化条件,使得本方法不仅能够得到准确的运动目标轮廓,而且满足了实时性要求。实验结果表明,该算法能够实时可靠地提取运动目标轮廓。Aiming at the problem that the moving target contour profile gotten by the motion detection algorithm is incomplete,inaccurate and so on,this paper presents a reliable extraction algorithm of moving objects based on the vibe algorithm and improved active contour model.In this article,the coarse position of moving targets is quiskly located with vible algorithm,combined with the improved active contour model,the real-time contour of moving targets can be extracted.The improved active contour model has good model initialization conditions.As to abtain the accurate contour of the moving object and meet the real-time requirement.The experimental results show that the algorithm can extract veliable and real-time contour of the moving object.

关 键 词：轮廓提取 Vibe算法 GVF SNAKE算法 轮廓粗定位 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]