基于连通检测和区域扩展的多重聚类方法研究  

Multiple clustering method based on connectivity detection and regional expansion

作　　者：钱涛[1] 马健[1] 刘凯[1] 党卫[1] 

机构地区：[1]中国人民解放军61541部队

出　　处：《航天电子对抗》2016年第5期36-39,共4页Aerospace Electronic Warfare

摘　　要：针对现有聚类算法在分选过程中存在辐射源定位和脉冲划分的困难,提出了基于连通检测和区域扩展的多重聚类方法。使用网格统计的方法对脉冲列分布密度进行量化,再借鉴图像分割算法,将量化的网格矩阵划分连通区域。基于峰值检测和梯度下降的原则扩展连通区域,通过边缘检测得到最终聚类结果。仿真结果表明,该算法能保证更多的聚类脉冲列参与参数分选,提高辐射源识别率。For the difficult of source positioning and pulse partition of existing clustering algorithms in the process of sorting, the clustering method based on the detection and regional expansion is put forward. Statistical methods use the grid to quantify the distribution density of the pulse train. By using image segmentation algorithm, quantization matrix grid is divided into connectivity region. The connectivity region is extended based on a peak detection and gradient descent principle, and the final result is given by the edge detection. Simulation results show that the clustering algorithm can ensure that more pulse train parameters involved in sorting, and inprove recognition rate.

关 键 词：圆概率误差 统计量化 连通检测 区域扩展 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]