多尺度样本熵在水面舰船识别中的应用研究  

AStudy of Surface Vessel Recognition Based on Multi- scale Sample Entropy

作　　者：刘磊[1] 鲍习中[1] 陈新华[1,2] 张志博[1] 孙长瑜[1] 

机构地区：[1]中国科学院声学研究所微弱信号检测与处理实验室,北京100190 [2]中国科学院水声环境特性重点实验室,北京100190

出　　处：《网络新媒体技术》2014年第5期28-31,共4页Network New Media Technology

基　　金：国家自然科学基金(61372180)

摘　　要：通过大量研究分析表明水面舰船的水声信号能够对其进行检测与识别并具有较好的性能,由于多尺度样本熵方法在离散信号分析中能够准确的定义其复杂性,本文采用多尺度样本熵方法分析水面舰船的水声信号。首先,利用多尺度样本熵方法计算不同水面舰船的水声信号。然后,将计算得到的多尺度样本熵值作为水面舰船特征,并采用支持向量机(SVM)分类器对水面舰船进行分类和识别。实验结果表明,本文提出的方法在水面舰船识别中具有较好的性能。A lot of research results reveal that the underwater acoustic signals include much information of the surface vessel, which could be used in the classification and recognition of the surface vessel. Moreover in view of the good performance of multi - scale sampie entropy method in definition of complexity of discrete signal, in this paper this method is adopted for the analysis of underwater acoustic signals of surface vessel. First, the values of underwater acoustic signals are calculated by multi - scale sample entropy method. Then, the calculated values which are viewed as the features of surface vessel are fed into support vector machine （SVM） classifier for the classification and recognition of the surface vessel. Experimental results show that the proposed method has good performance in the classification and recognition of the surface vessel.

关 键 词：水面舰船识别 特征提取 多尺度样本熵 

分 类 号：U675.7[交通运输工程—船舶及航道工程] TN912.3[交通运输工程—船舶与海洋工程]