利用小波和分形理论进行水下回波的特征提取
Feature extraction of underwater echoes using wavelet and fractal theories
-
摘要: 该文首先分析了五类湖底回波的不同尺度下小波子空间的能量特征和分形维特征；然后将这些特征矢量作为分类的特征，并根据特征本身的离散程度对其进行加权；最后采用最小距离分类器对其进行分类，取得了 96.11％的分类正确率。
-
关键词:
- 小波变换; 分形; 特征提取
Abstract: Firstly, the energy distribution in different wavelet scale space and the fractal dimension of underwater sonar echoes are discussed. Then, these feature vectors are utilized to classify the real echoes, and weight these features according to their own degree of dispersion. Finally, a minimum distance classifier is used in the classification procedure, and experimental results demonstrate the efficiency of the method. -
王正垠,马远良,D.Vray,等,湖底回波信号的解卷积处理[J],信号处理,1997,13(4),335-340.[2]赵建平,黄建国,谢一清,等,用小波变换进行水下回波边缘特征提取与分类识别[J],声学学报,1998,23(1),31-38.[3]张静远,张冰,蒋兴舟,基于小波变换的特征提取方法研究[J],信号处理,2000,16(2),156-162.[4]丁庆海,庄志洪,祝龙石,等,混沌、分形和小波理论在被动声信号特征提取中的应用[J],声学学报,1999,24(2),197-203.[5]余秋星,基于混沌与分形的信号检测[M],[硕士论文],西安,西北工业大学,2001.[6]S.Haykin,XB Li,Detection of Signals in Chaos,Proc.IEEE[J], 1995,83(1),94-122.[7]吴高洪,章毓晋,林行刚,利用特征加权对基于小波变换的纹理分类[J],模式识别与人工智能,1999,12(3),262-267. -
计量
- 文章访问数: 2018
- HTML全文浏览量: 110
- PDF下载量: 808
- 被引次数: 0
下载:
