Because of the complexity of underwater environment, detecting and recognizing of underwater target is a difficult problem in the area of underwater acoustic signal processing. A method based on wavelet transform ( WT) and probabilistic neural network ( PNN ) for underwater target recognition is studied in the paper. Energy distribution at different scales of radiated noise of underwater target is derived by WT with which severed as feature vectors. Then the feature vectors act as input vectors of PNN for target classification. The energy vectors by WT can differentiate various targets' radiated noise. The design of PNN is straightforward and does not depend on training. PNN is suitable for signal classification. The result from test shows that the method is effective and feasible.%由于水下环境的复杂性,水下目标的检测和识别是水声信号处理领域中的一个难题.本文研究了基于小波变换和概率神经网络的水下目标识别方法.利用小波变换得到水下目标辐射噪声信号在不同尺度下的能量分布作为特征矢量,并输入到概率神经网络中以实现目标分类.利用小波变换能量特征值可有效区分不同的目标辐射噪声.概率神经网络无网络训练过程,适合于信号分类.实验结果表明该方法的有效性和可行性.
展开▼