Time-scale wavelet scattering using hyperbolic tangent function for vessel sound classification
Date
2017-08-09
Authors
Editor(s)
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
Source Title
25th European Signal Processing Conference, EUSIPCO 2017
Print ISSN
Electronic ISSN
Publisher
IEEE
Volume
Issue
Pages
1794 - 1798
Language
English
Type
Journal Title
Journal ISSN
Volume Title
Series
Abstract
We introduce a time-frequency scattering method using hyperbolic tangent function for vessel sound classification. The sound data is wavelet transformed using a two channel filter-bank and filter-bank outputs are scattered using tanh function. A feature vector similar to mel-scale cepstrum is obtained after a wavelet packed transform-like structure approximating the mel-frequency scale. Feature vectors of vessel sounds are classified using a support vector machine (SVM). Experimental results are presented and the new feature extraction method produces better classification results than the ordinary Mel-Frequency Cepstral Coefficients (MFCC) vectors. © EURASIP 2017.
Course
Other identifiers
Book Title
Keywords
Hyperbolic tangent function , Scattering filter-bank , Timefrequency representation , Vessel sound classification , Classification (of information) , Filter banks , Hyperbolic functions , Image retrieval , Signal processing , Speech recognition , Support vector machines , Vectors , Classification results , Feature extraction methods , Hyperbolic tangent function , Mel frequencies , Mel-frequency cepstral coefficients , Sound classification , Time-frequency representations , Two-channel filter banks , Acoustic wave scattering