[1]曹红丽,方世良.舰船辐射噪声的响度和音色特征模型[J].东南大学学报(自然科学版),2013,43(2):241-246.[doi:10.3969/j.issn.1001-0505.2013.02.003]
 Cao Hongli,Fang Shiliang.Loudness and timbre features model of ship radiated noise[J].Journal of Southeast University (Natural Science Edition),2013,43(2):241-246.[doi:10.3969/j.issn.1001-0505.2013.02.003]
点击复制

舰船辐射噪声的响度和音色特征模型()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
43
期数:
2013年第2期
页码:
241-246
栏目:
信息与通信工程
出版日期:
2013-03-20

文章信息/Info

Title:
Loudness and timbre features model of ship radiated noise
作者:
曹红丽方世良
东南大学水声信号处理教育部重点实验室, 南京 210096
Author(s):
Cao Hongli Fang Shiliang
Key Laboratory of Underwater Acoustic Signal Processing of Ministry of Education, Southeast University, Nanjing 210096, China
关键词:
Moore响度 谱质心 谱质心带宽 谱质心斜度 谱质心峰度 BP神经网络
Keywords:
Moore loudness spectral centroid spectral centroid bandwidth spectral centroid skewness spectral centroid kurtosis BP neural network
分类号:
TN911.7
DOI:
10.3969/j.issn.1001-0505.2013.02.003
摘要:
首先分析模拟舰船辐射噪声的归一化功率和Moore响度变化趋势,进一步用海试数据验证Moore响度的周期性变化特点,分析Moore响度周期性变化的原因,提取Moore响度均值作为一维特征.然后分析研究音色中有效的谱质心特征,根据谱质心的数学定义推导出谱质心带宽、谱质心斜度以及谱质心峰度的数学公式,举例分析其物理含义,并提出基于六分段频带的谱质心、谱质心带宽、谱质心斜度以及谱质心峰度共24维的音色特征模型.将Moore响度特征模型和音色特征模型相结合,从组合的25维特征的均值和方差分布可知特征具有一定的可区分性.最后设计3层BP神经网络分类器,海试数据表明基于Moore响度和音色的特征模型提取方法是有效的,且有利于提高分类识别率.
Abstract:
First, the normalized power and Moore loudness change trend of the simulated ship radiated noise is analyzed. Furthermore, the Moore loudness cyclical change is tested by the sea trial data, and the Moore loudness mean is extracted as one dimensional characteristic. Then the effective timbre spectrum centroid characteristics are studied, and the formulae of spectral centroid bandwidth, spectral centroid skewness and spectral centroid kurtosis are deduced according to the mathematical definition of spectrum centroid. The physical meanings of spectrum centroid, spectral centroid bandwidth, spectral centroid skewness and spectral centroid kurtosis are illustrated by an example, and a 24-dimensional features model based on spectral centroid, spectral centroid bandwidth, spectral centroid skewness and spectral centroid kurtosis of six segmented frequency bands is proposed. The Moore loudness feature model is combined with the timbre feature model, and the combined characteristics have certain separability, which can be seen from the mean and variance distributions of 25-dimensional characteristics. Finally, a three-layer BP neural network classifier is designed, and the sea trial data simulation results indicate that the feature extraction method is effective and helpful to improve the recognition rate.

参考文献/References:

[1] 刘载芳, 王大训, 张友奎. 声纳听音判型 [M]. 北京: 海军出版社,1999.
[2] Park T H. Automatic musical instrument recognition [D]. Tampere, Finland: Tampere University of Technology, 2001.
[3] Giulio A, Maurizio L, Emanuele P. Musical instrument timbres classification with spectral features [J]. Journal on Advance in Signal Processing, 2003, 2003(1): 5-14.
[4] Xin Z, Zbigniew W. Analysis of sound features for musical timbre recognition [C]//International Conference on Multimedia and Ubiquitous Engineering. Seoul,Korea, 2007: 26-28.
[5] 王娜, 陈可安. 水下噪声音色属性回归模型以及在目标识别中的应用[J]. 物理学报, 2010, 59(4): 2873-2881.
  Wang Na, Chen Ke’an. Regression model of timbre attribute for underwater noise and its application to target recognition [J]. Acta Physica Sinica, 2010, 59(4): 2873-2881.(in Chinese)
[6] Zwicker E, Fastl H. Psychoacoustics: facts and models [M]. 2nd ed. Berlin: Spring-Verlag, 1999: 204-237.
[7] Moore B C J, Glasberg B R,Baer T. A model for the prediction of thresholds, loudness, and partial loudness [J]. Acoustical Society of America, 1997, 45(4): 224-239.
[8] Moore B C J. ANSE S3.4-2005 Procedure for the computation of loudness of steady sounds [S]. New York: American National Standards Institute, 2005.
[9] 王娜, 陈克安. 分段谱质心特征在水下目标识别中的应用[J]. 兵工学报, 2009, 30(2): 144-149.
  Wang Na, Chen Ke’an. Application of sub-band spectral centroid features to recognition underwater targets [J]. Acta Armamentarii, 2009, 30(2): 144-149.(in Chinese)
[10] 钱晓南. 舰船螺旋桨噪声 [M]. 上海:上海交通大学出版社, 2011.

备注/Memo

备注/Memo:
作者简介: 曹红丽(1985—),女,博士生;方世良(联系人),男,博士,教授,博士生导师,slfang@seu.edu.cn.
基金项目: 国家重大基础研究发展计划(973计划)资助项目(6131222)、国家自然科学基金资助项目(11104029).
引文格式: 曹红丽,方世良.舰船辐射噪声的响度和音色特征模型[J].东南大学学报:自然科学版,2013,43(2):241-246. [doi:10.3969/j.issn.1001-0505.2013.02.003]
更新日期/Last Update: 2013-03-20