基于模糊聚类的半自动跟踪技术研究

A New Semi-automatic Tracking Method Based on Fuzzy Clustering Algorithm

现有半自动跟踪系统将操作员等效为准线性环节,增大了训练的难度和成本,不适应复杂战场环境下多个快速机动目标的选取和跟踪。利用特征提取方法得到视场的"潜在目标"集,引入模糊聚类方法,通过对操作响应进行分类,建立操控意图与目标运动特征之间的关系模型。通过充分的仿真试验,对比研究了跟踪平稳性、快速性、准确性等指标,结果说明该方法能够有效提升整体跟踪性能。

The current semi-automatic tracking system takes the operator as approximate linear process. This enlarges the cost of human training and the training difficulty. It couldn＇t suit for the complex war condition. This paper puts fuzzy clustering algorithm into tracking system, gets the potential goals by image processing and classifies the response of the operation, thus the relationship between user purpose and character of targets is made up. With a great deal of experiment, the velocity, the stabilization and the precision are analyzed, and the results show that the method could improve the performance of target tracking obviously.