|Title:||Automatic classification of pinnipeds based on their vocalizations and fusion of cepstral features||Authors:||Noda, J. J.
Travieso, Carlos M.
Dutta, M. K.
|UNESCO Clasification:||33 Ciencias tecnológicas
3307 Tecnología electrónica
Pinniped's call recognition
|Issue Date:||2016||Conference:||3rd International Conference on Signal Processing and Integrated Networks, SPIN 2016||Abstract:||Acoustic vocalizations are common in marine mammals which can be used for classification purposes. Pinnipeds are a group of carnivore mammals composed by seals, sea lions, and walruses. But although, there is a great interest in research literature about acoustic monitoring of marine mammals, the identification of pinnipeds trough experts systems has been poorly studied. This paper brings a novel method for the automatic taxonomic classification of pinnipeds using a fusion of Mel and Linear Frequency Cepstral Coefficients (MFCCs and LFCCs), representing the acoustic signal in both low and high frequencies. In our experiment, we have used kNN for classification, achieving an average identification of 96.48% ± 9.17 over 18 pinniped species.||URI:||http://hdl.handle.net/10553/42442||ISBN:||9781467391979||DOI:||10.1109/SPIN.2016.7566691||Source:||2016 3Rd International Conference On Signal Processing And Integrated Networks (Spin), p. 215-219|
|Appears in Collections:||Actas de congresos|
Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.