|
|
论文:2014,Vol:32,Issue(6):906-911 |
|
|
引用本文: |
|
|
吴姚振, 杨益新, 田丰, 杨龙, 陶灿. 基于Gammatone频率离散小波系数的水下目标鲁棒识别[J]. 西北工业大学学报 |
|
|
Wu Yaozhen, Yang Yixin, Tian Feng, Yang Long, Tao Can. Robust Underwater Target Recognition Based on Gammatone Frequency Discrete Wavelet Coefficients (GFDWC)[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
基于Gammatone频率离散小波系数的水下目标鲁棒识别 |
|
吴姚振, 杨益新, 田丰, 杨龙, 陶灿 |
|
西北工业大学 航海学院, 陕西 西安 710072 |
摘要: |
针对水下目标辐射噪声的复杂性和研究样本的局限性,提出了一种基于Gammatone频率离散小波系数的特征提取方法,结合人耳听觉感知机理,提取出了有效吻合人耳听觉特性的识别特征。该方法在2个方面改进了目前广泛采用的美尔倒谱系数:1用Gammatone滤波器代替三角滤波器,更好地模拟了人耳基底膜的滤波特性;2用离散小波变换替换离散余弦变换,使得识别特征具有优良的局部化特性。针对实录的水下动物叫声和舰船辐射噪声进行分类实验,表明所提出的特征提取方法在识别率和稳健性方面都有明显提高。 |
关键词:
Gammatone滤波器
Gammatone频率离散小波系数
水下目标识别
特征提取
|
|
Robust Underwater Target Recognition Based on Gammatone Frequency Discrete Wavelet Coefficients (GFDWC) |
|
Wu Yaozhen, Yang Yixin, Tian Feng, Yang Long, Tao Can |
|
School of Marine Science and Technology, Northwestern Polytechnical University, Xi'an 710072, China |
Abstract: |
In this paper,aiming at the complexity of underwater target-radiated noise and the limitations of studysamples,we present a novel recognition feature approach based on GFDWC,which imitates the hearing perceptionmechanism of humans. Compared with Mel-frequency cepstral coefficients (MFCC) used extensively at present,this method improves the performance of underwater target recognition system in the following two aspects: (1) thetriangular filters were improved based on gammatone function,which approximates well the filter response of basilarmembrane; (2) taking localization characteristics into account,we replaced the discrete cosine transform (DCT)with discrete wavelet transform (DWT). The results of the recognition experiments of real underwater target-radiatednoise and their analysis show preliminarily that the proposed method has good robustness and gives high recognitionrates. |
Key words:
gammatone filter
gammatone frequency discrete wavelet coefficients (GFDWC)
underwater targetrecognition
feature extraction
discrete
cosine transforms
disarete wavelet transforms
multiresolu-tion unalysis
|
|
收稿日期: 2014-05-16
修回日期:
|
DOI: |
基金项目: 国家自然基金(11274253)资助 |
通讯作者:
Email: |
作者简介: 吴姚振(1985-),西北工业大学博士研究生,主要从事水声信号处理与目标识别研究。
|
|
相关功能 |
|
|
|
作者相关文章 |
|
吴姚振 在本刊中的所有文章 |
杨益新 在本刊中的所有文章 |
田丰 在本刊中的所有文章 |
杨龙 在本刊中的所有文章 |
陶灿 在本刊中的所有文章 |
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Tucker S,Brown G J. Classification of Transient Sonar Sounds Using Perceptually Motivated Features [J]. IEEE Journal ofOceanic Engineering,2005,30(3): 588-600 [2] 彭圆,王晟,王科俊,等. 感知线性预测在水下目标分类中的应用研究[J]. 声学学报,2006,31(2): 146-150Peng Yuan,Wang Cheng,Wang Kejun,et al. A Study on Underwater Target Classification Applying Perception Linear Predic-tion Method[J]. Acta Acustica,2006,31(2): 146-150 (in Chinese) [3] 马元锋,陈克安,王娜,等. 听觉模型输出谱特征在声目标识别中的应用[J]. 声学学报,2009,34(2): 142-150Ma Yuanfeng,Chen Kean,Wang Na,et al. Application of Auditory Spectrum-Based Features into Acoustic Target Recognition[J]. Acta Acustica,2009,34(2): 142-150 (in Chinese) [4] Valero X,Alias F. Gammatone Cepstral Coefficients: Biologically Inspired Feature for Non-Speech Audio Classification [J].IEEE Trans on Multimedia,2012,14(6): 1684-1689 [5] Wang X Y. The Improved MFCC Speech Feature Extraction Method and Its Application [J]. Advanced Materials Research,2013,756: 4059-4062 [6] Wan W G,Au O C. Robust Speech Recognition Based on the Second-Order Difference Cochlear Model [C] ∥Proceedings ofIMVSP,2001,543-546 [7] Smith J O,Abel J S. Bark and ERB Bilinear Trnasform [J]. IEEE Trans on Speech and Audio Processing,1999,7(6):697-708 [8] Lyon R F,Katsiamis A G,Drakakis. History and Feature of Auditory Filter Models[C] ∥Proceedings of IEEE ISCAS,2010:3809-3812 [9] Lemonds D W,Au W W L,Vlachos S A,et al. High-Frequency Auditory Filter Shape for the Atlantic Bottlenose Dolphin[J].J Acout Soc Am,2012,132(2): 1222-1228 [10] Strahl S,Mertins A. Analysis and Design of Gammatone Signal Models [J]. J Acout Soc Am,2009,126(5): 2379-2389 [11] Wu Y Z,Yang Y X,Tao C,et al. A Study on Underwater Target Recognition Applying Auditory Slow Feature Analysis[C] ∥Proceedings of OCEANS'14 MTS/IEEE,2014: 1-5 [12] Irino T,Aoki Y,Kawahara H,et al. Comparison of Perfoemance with Voiced and Whispered Speech in Word Recognition andMean-Format-Frequency Discrimination[J]. Speech Communication,2012,54: 998-1013 [13] Jin X,Gupta S,Mukherjee K,et al. Wavelet-Based Feature Extraction Using Probabilistic Finite State Automata for PatternClassification[J]. Pattern Recognition,2010,44: 1343-1356 |
|
|
|
相关文献: |
|
|
1.吴姚振, 杨益新, 杨龙, 王永威.基于恒定束宽波形保真及干扰抑制的水下目标识别方法[J]. 西北工业大学学报, 2015,33(5): 843-848 |
|
|
|
|
|
|