摘要
对海战场空间观察对象中的目标分群问题进行了描述;定义观察目标航迹时空距离计算模型及聚类算法,与依据观察目标位置、航向、航速特征量构建的相似度度量方法的传统目标分群层次聚类算法进行了分析对比。该算法将观察目标的时态信息与空间信息相结合,提高了海战场目标分群聚类的质量。
Target grouping is one of the main task of sea battlefield situation assessment.The distance between target trajectories and cluster algorithm are presented.The algorithm makes use of spatio-temporal data.Compared with the traditional hierarchy algorithm which uses the characteristic measure of location,speed and direction of targets.The advantage of this method received much higher recognition accuracy rate.
出处
《计算机与数字工程》
2010年第5期28-30,48,共4页
Computer & Digital Engineering
关键词
目标分群
航迹近邻
距离尺度
target grouping
trajectories near neighbor
distance scales