摘要
针对水下监控系统中的小目标跟踪问题,在分析蛙人探测声纳图像特点的基础上,提出了基于面积约束的全局域最近邻数据关联法,有效解决了水下监控环境下目标小、结构信息缺乏、成像不稳定的难点,通过引入航迹"运动方向"的思想消除了混响区虚假航迹的影响,实验证明该算法具有准确性高、实时性好、鲁棒性强的特点。
To resolve the problem of small targets tracking in underwater surveillance system,an area-constrained global nearest neighbour data association method is proposed based on the analysis of the features of diver detection sonar images. It overcomesthe difficulties of targets being small,instability and the absence of structure information under underwater low SNR environment,resolves the problem of false tracks in reverberation region by the ideas of"movement direction". Experiments show it is a practicalmethod that is robust,fast in speed and good in veracity.
引文
[1]A.Asada,K.Kuramoto,T.Tanaka,K.Oimatsu,et al.Devel-opment of an Underwater Security Sonar System[J].pro-ceedings of Oceans'06,2006:1-3.
[2]Crawford A.M,Vance Crowe D.Observations from Demon-strations of Several Commercial Diver Detection Sonar Sys-tems[J].proceedings of Oceans'07,2007:1-3.
[3]Annati M.Harbour protection and defense of the coast[J].Naval Forces 2008,29(III):24-37.
[4]蒋立军,杨娟,许枫.蛙人探测声呐技术研究进展[J].中国科学,2009,54(3):269-272.
[5]张波,陈小泉.蛙人探测及蛙人探测声呐[J].中国科技信息,2009,2:51-52.
[6]H.Bohyung,D.Comanuciu,Z.Ying,D.Larry S.Sequentialkernel density approximation and its application to re-al-time visual tracking[J].IEEE Trans.On Pattern rec-ognition and machine intelligence,2008,30(7):1186-1197.
[7]侯志强,韩崇昭.视觉跟踪技术综述[J].自动化学报,2006,32(4):603-617.
[8]Bell J.M,Petillot Y.R,Lebart K,et al.Target Recognition in Synthetic Aperture and High Resolution Sidescan Sonar[J].High Resolution Imaging and Target Classification,2006,11:99-106.
[9]Clark D,Ruiz Z.T,Petillot Y,Bell J.Particle PHD filter multiple target tracking in sonar images[J].IEEE Trans.On Aerospace and Electronic System 2007,43(1):409-416.
[10]蒋立军,杨娟,许枫.蛙人探测声纳技术研究进展[J].科学通报,2009,54(3):269-272.