一种基于小波分频带统计特征的舰船分类识别方法  

An Identification Method of Ship Classification Based on Wavelet Frequency Band Statistical Features

作　　者：潘琳[1] 张效民[1] 刘义海[1] 

机构地区：[1]西北工业大学航海学院,陕西西安710072

出　　处：《鱼雷技术》2013年第1期76-80,共5页Torpedo Technology

摘　　要：为了有效提高舰船目标的识别率,提出了一种基于小波分频带统计特征的舰船分类识别方法。该方法利用小波变换分别提取了舰船辐射噪声带内信号的小波模极大值和带外信号的分频段能量两类特征,并将两类目标特征的联合量作为舰船的目标特征用以识别舰船目标。同时利用实测海录58组舰船的辐射噪声数据对上述舰船目标特征的分类识别方法进行了验证,结果表明,利用联合特征对目标的检测概率可以达到97%以上。This paper proposes an identification method of ship classification based on wavelet frequency band statisti-cal features to improve effectively the detection probability of a ship. In this method, the maximum of wavelet modulus for in-band signal of ship-radiated noise and the frequency band energy of outer band signal are extracted, respectively,by using wavelet transform, and the combination of the two target features is used for identifying a ship target. This method is verified by 58 sets of measured data of radiated noise from sea trial, and the results show an improved detec- tion probability of 97%.

关 键 词：舰船辐射噪声 小波变换 特征提取 目标分类识别 

分 类 号：TJ630.34[兵器科学与技术—武器系统与运用工程] TB533.2[理学—物理]