Problems about target classification in situation assessment are analyzed. This paper presents a syntheticmethod for fulfilling target classification by using the nearest-neighbor method and field knowledge. The graph...Problems about target classification in situation assessment are analyzed. This paper presents a syntheticmethod for fulfilling target classification by using the nearest-neighbor method and field knowledge. The graphicalstructure formed by target classification is shown by the adjacency list. Based on the structure, breadth-first searchalgorithm is used for the implementation of dynamic maintenance. The output of target classification is helpful to de-termine the interaction among situation elements, thus interprets actions related to problem field.展开更多
针对目标编群中单一算法存在的适用范围小、误分率高的问题,提出一种新的态势估计中目标编群的处理方法。首先应用Hop fie ld神经网络对态势中目标的目的地做出判断,然后采用多相似性加权策略计算出目标间的相关系数,再根据最大相关系...针对目标编群中单一算法存在的适用范围小、误分率高的问题,提出一种新的态势估计中目标编群的处理方法。首先应用Hop fie ld神经网络对态势中目标的目的地做出判断,然后采用多相似性加权策略计算出目标间的相关系数,再根据最大相关系数层次聚类算法实现编群。仿真结果表明方法能在一定程度上减小错误编群的概率,同时适用范围也得到了扩展。展开更多
文摘Problems about target classification in situation assessment are analyzed. This paper presents a syntheticmethod for fulfilling target classification by using the nearest-neighbor method and field knowledge. The graphicalstructure formed by target classification is shown by the adjacency list. Based on the structure, breadth-first searchalgorithm is used for the implementation of dynamic maintenance. The output of target classification is helpful to de-termine the interaction among situation elements, thus interprets actions related to problem field.
文摘针对目标编群中单一算法存在的适用范围小、误分率高的问题,提出一种新的态势估计中目标编群的处理方法。首先应用Hop fie ld神经网络对态势中目标的目的地做出判断,然后采用多相似性加权策略计算出目标间的相关系数,再根据最大相关系数层次聚类算法实现编群。仿真结果表明方法能在一定程度上减小错误编群的概率,同时适用范围也得到了扩展。
