基于小波变换和支持向量机的水下目标分类方法

水下目标识别技术是世界各国十分重视的研究课题之一,具有重要的理论和应用价值。简要介绍了小波变换、多分辨率分析及支持向量机的基本原理,针对舰船辐射噪声信号,利用小波变换来完成信号的预处理和滤波,在小波变换后信号的多尺度子空间上提取信号的能量特征参数,归一化处理后构建特征向量,最后用支持向量机算法进行分类。仿真结果表明,利用小波变换的多分辨率分析和支持向量机能对舰船辐射噪声信号进行有效分类识别。

The Method of Underwater Target Classification Based on Wavelet Transform and Support Vector Machine

Underwater target recognition,which possesses a crucial theoretic and application value,is a very important topic in the world.The basic principles of Wavelet Transform(WT),Multi-Resolution Analysis(MRA) and Support Vector Machine(SVM) are briefly introduced at first.Then the signals of ship radiated are preprocessed and filtered by WT and the energy feature parameters are extracted upon multi-scale subspace after WT.At last,the characteristic vector is established with normalization process and the signals...