基于联合双边滤波的融合图像跟踪算法研究  

A Tracking Algorithm for Fusion Image Based on Joint Bilateral Filter

作　　者：王跃东[1] 钱小燕[1] 张天慈[1] 孙泽军[1] 

机构地区：[1]南京航空航天大学民航学院,江苏南京210016

出　　处：《航空计算技术》2016年第2期37-41,共5页Aeronautical Computing Technique

基　　金：中国博士后基金项目资助(20110491415);南京航空航天大学青年科学基金项目资助(NN2012049);南京航空航天大学研究生创新基地(实验室)开放基金(kfjj20150707)

摘　　要：为了提高跟踪精度,提出一种基于联合双边滤波的融合图像自动跟踪算法。采用帧间差分法并通过形态学处理获得运动目标大致区域,在此基础上采集目标区域及相邻背景区域的代表颜色集合,由此生成草图作为联合双边滤波的引导图,进而生成完整的目标描述,通过边缘检测获得精确的跟踪结果。实验结果表明,方法能够获得清晰准确的跟踪结果,能够较好地避免跟踪中的漂移问题,且算法复杂度低,跟踪精度高。An precise object tracking algorithm for fusion image based on joint bilateral fiher was proposed in this paper to improve the tracking effect. Firstly, frame difference and the morphology processing are applied to produce the coarse target, which helps to collect the representative color of target area and the region of the background. Then a sketch will be generated as the guide map of the joint bilateral filter, which helps to produce the complete target description. Finally through edge detection, we can obtain the accurate tracking results. Extensive experiments show that the proposed method can obtain clear and ac- curate tracking results, which can avoid the drift problem in tracking. At the same time, this algorithm has lower complexity and higher tracking accuracy.

关 键 词：联合双边滤波 目标跟踪 融合图像 引导图像 

分 类 号：O242[理学—计算数学]