水下不同材料的回波特征提取与分类研究
详细信息 查看官网全文
- 英文论文题名：Echo feature extraction and classification of different underwater materials
- 论文作者：金磊磊 ; 梁红 ; 陆禹
- 英文论文作者：JIN Lei-lei ; LIANG Hong ; LU Yu ; Northwestern Polytechnical University ; College of Marine Science and Technology
- 年：2016
- 作者机构：西北工业大学航海学院;
- 论文关键词：水下目标 ; 材料分类 ; 小波变换 ; 倒谱分析 ; 神经网络
- 英文论文关键词：underwater target ; materials classification ; wavelet transform ; cepstral analysis ; neural network
- 会议召开时间：2016-08-21
- 会议录名称：2016’中国西部声学学术交流会论文集
- 语种：中文
- 分类号：TB56
- 学会代码：SXJS
- 会议名称：2016’中国西部声学学术交流会
- 会议地点：中国四川成都
- 主办单位：中国声学学会微声学分会、中国声学学会物理声学分会、陕西省声学学会、浙江省声学学会、山东省声学学会、上海市声学学会、黑龙江省声学学会、水声技术重点实验室、重庆市声学学会、西安市声学学会、四川省声学学会
- 学会名称：《声学技术》编辑部
- 页数：4
- 文件大小：471k
- 原文格式：D
- 会议级别：全国
摘要
水下目标的分类识别在国家海上防卫和海洋科学考察等方面的应用日益广泛。主动声呐的回波是入射声波作用于目标后产生的一种物理过程,其中包含了大量的目标特征信息,与目标的材料等属性有关。根据不同材料回波的差异对材料进行分类,有助于复杂水下环境中对物体的进一步识别。考虑到小波变换的多分辨分析能力以及倒谱分析在被动声呐目标识别中的成功应用,分别提取回波信号的尺度-小波能量谱特征和反映材料冲激响应特性的倒谱特征。设计人工神经网络分类器,利用从实测数据提取的上述两种特征对3类材料(铁板、铝板、石砖)进行分类。分析比较利用不同特征量及结合两种特征量的分类结果,验证了基于小波变换特征和倒谱特征的水下材料分类的可行性。
Classification and recognition of underwater target in such aspects as national maritime defense and marine scientific research are becoming more and more widely used. Active sonar echo signal is a physical process generated after the incident signal acts on the target, which contains various target feature information, associated with the material property of the target. Classifying materials by their different echoes can contribute to further recognition of objects in complex underwater environment. Considering the multi-resolution analysis of wavelet transform and the successful application of cepstrum analysis in passive sonar target recognition, scale-wavelet power spectrum characteristic and cepstrum characteristic are extracted respectively from the echo signal. Artificial neural network classifier is designed, using characteristics extracted from the experimental data of three kinds of materials(iron, aluminum, stone) to classify them. Classification results by using different characteristic are analyzed and compared, showing the validity of the two kinds of characteristics mentioned above in the application of underwater material classification.
Classification and recognition of underwater target in such aspects as national maritime defense and marine scientific research are becoming more and more widely used. Active sonar echo signal is a physical process generated after the incident signal acts on the target, which contains various target feature information, associated with the material property of the target. Classifying materials by their different echoes can contribute to further recognition of objects in complex underwater environment. Considering the multi-resolution analysis of wavelet transform and the successful application of cepstrum analysis in passive sonar target recognition, scale-wavelet power spectrum characteristic and cepstrum characteristic are extracted respectively from the echo signal. Artificial neural network classifier is designed, using characteristics extracted from the experimental data of three kinds of materials(iron, aluminum, stone) to classify them. Classification results by using different characteristic are analyzed and compared, showing the validity of the two kinds of characteristics mentioned above in the application of underwater material classification.
引文
[1]汤渭霖.声呐目标回波的亮点模型[J].声学学报,1994,19(2):93-100.TANG Weilin.Highlight model of echoes from sonar targets[J].Acta Acustica,1994,19(2):93-100.
[2]马艳,李志舜.基于连续小波变换的水下目标特征提取与分类[J].系统工程与电子技术,2003,25(3):375-378.Ma Yan,Li Zhishun.Feature Extraction and Classification of Underwater Target Based on CWT[J].Systems Engineering and Electronics,2003,25(3):375-378.
[3]Tufts D W,Ianniello J P,Lourtie I,et al.The past,present and future of underwater acoustic signal processing[J].IEEE Signal Processing Magazine,1998,15(4):22-51.
[4]柳革命,孙超,杨益新.两种倒谱特征提取技术在水声目标识别中的应用[J].西北工业大学学报,2008,26(3):276-281.Liu Geming,Sun Chao,Yang Yixin.Feature Extraction of Passive Sonar Target Based on Two Cepstrums.Journal of Northwestern Polytechnical University,2008,26(3):276-281.
[5]Ren C,An N,Wang J,et al.Optimal parameters selection for BP neural network based on particle swarm optimization:A case study of wind speed forecasting[J].Knowledge-Based Systems,2013,56(3):226-239.
[2]马艳,李志舜.基于连续小波变换的水下目标特征提取与分类[J].系统工程与电子技术,2003,25(3):375-378.Ma Yan,Li Zhishun.Feature Extraction and Classification of Underwater Target Based on CWT[J].Systems Engineering and Electronics,2003,25(3):375-378.
[3]Tufts D W,Ianniello J P,Lourtie I,et al.The past,present and future of underwater acoustic signal processing[J].IEEE Signal Processing Magazine,1998,15(4):22-51.
[4]柳革命,孙超,杨益新.两种倒谱特征提取技术在水声目标识别中的应用[J].西北工业大学学报,2008,26(3):276-281.Liu Geming,Sun Chao,Yang Yixin.Feature Extraction of Passive Sonar Target Based on Two Cepstrums.Journal of Northwestern Polytechnical University,2008,26(3):276-281.
[5]Ren C,An N,Wang J,et al.Optimal parameters selection for BP neural network based on particle swarm optimization:A case study of wind speed forecasting[J].Knowledge-Based Systems,2013,56(3):226-239.