基于邻域拓扑结构的目标协同跟踪算法-计算机工程与应用.PDFVIP

Computer Engineering and Applications 计算机工程与应用 2016 ，52（7 ） 101 基于邻域拓扑结构的目标协同跟踪算法 程雪雪，陈 莹 CHENG Xuexue, CHEN Ying 江南大学 轻工过程先进控制教育部重点实验室，江苏 无锡 214000 Key Laboratory of Advanced Process Control for Light Industry （Ministry of Education ）, Jiangnan University, Wuxi, Jiangsu 214000, China CHENG Xuexue, CHEN Ying. Collaborative tracking algorithm based on target neighborhood topology. Computer Engineering and Applications, 2016, 52 （7 ）：101-105. Abstract ：Aiming at the problem of target tracking, a new algorithm for cooperative target monitoring and tracking is pro- posed based on neighborhood topology of nodes. According to communication differences when nodes share information, the approach calculates the capture gain of each node based on communication probability, through which the cluster members are determined and the precision of targets position is improved. Finally, the final area of the target is defined by the intersection of the nodes sensing area and the maximum range of target movement. The area centroid is calculated using the grid method. The simulation results show that the proposed approach outperforms other similar methods in both tracking accuracy and stability. Key words ：target tracking; cooperative; neighborhood topology; communication probability; capture gain 摘 要：针对目标跟踪问题，提出一种新的节点协同跟踪方法，通过节点间邻域拓扑结构协同建立各节点的关联性， 对目标进行协作监测与跟踪。算法根据节点信息共享时的传播差异，通过节点间的传播概率准确计算各节点捕捉 增益，从而确定簇成员，增加了确定目标位置的精度。结合感知区域及最大移动定理界定目标最终移动区域，采用 网格法算出区域质心。实验结果证明，与同类方法相比，该算法具有较高的跟踪精度，且跟踪稳定度较好。 关键词：目标跟踪；协同；邻域拓扑结构；传播概率；捕捉增益 文献标志码：A 中图分类号：TP391 doi ：10.3778/j.issn. 1002-8331.1405-0040 1 引言 点能量消耗较少，节点信息随之减少，跟踪精度也难以保 协同跟踪是资源受限视频传感器网络中一种比较 证，所以有限的节点资源如何有效相互协作则显得尤为 常见的处理方法，该技术目的在于协同多个视频传感器 重要，那么如何找到一种协作跟踪算法成了研究人员关 [1] 注的热点。Yuan[5]将场景