摘要
小波变换是处理非平稳信号的一个有力工具,研究了基于小波包分析的声信号特征提取方法,并应用该方法对直升机等4种目标的声信号进行了特征提取,降低了特征向量的维数。采用设计改进的BP神经网络分类器对声目标进行分类,分类结果准确率高,获得满意的实验效果。
Wavelet theory is a useful tool for processing the non -placement signal, this paper analyzes the feature extraction method of battlefield acoustic signal based on the wavelet packet analysis theory. Features of four types of acoustic signals of the battlefield target are extracted and low - dimension feature vectors are obtained by using this technique. BP neural network classifier is designed for the acoustic target classification. Satisfactory experimental results are obtained with highly classification accuracy.
出处
《空军工程大学学报（自然科学版）》
CSCD
北大核心
2007年第6期40-43,共4页
Journal of Air Force Engineering University(Natural Science Edition)
基金
陕西省自然科学研究项目(2004F36)
关键词
小波包分析
特征提取
分类器
目标识别
wavelet packet analysis
feature extraction
classifier
target identification
