Fernando Merchan, a), b) Giacomo Echevers, and Hector Poveda) Facultad de Ingeniería Eléctrica, Universidad Tecnologica de Panama, P.O. Box 0819-07289, El Dorado, Panama, Republic of Panama.
Javier E. Sanchez-Galana), c) Facultad de Ingenierıa de Sistemas Computacionales, Universidad Tecnologica de Panama, P.O. Box 0819-07289, El Dorado, Panama, Republic of Panama.
Hector M. Guzman Smithsonian Tropical Research Institute, P.O. Box 0843-03092, Panama, Republic of Panama.
Received 27 October 2018; revised 15 August 2019; accepted 24 August 2019; published online 25 September 2019
This work presents a methodology to automatically detect and identify manatee vocalizations in continuous passive acoustic underwater recordings. Given that vocalizations of each manatee present a slightly different frequency content, it is possible to identify individuals using a noninvasive acoustic approach. The recordings are processed in four stages, including detection, denoising, classification, and manatee counting and identification by vocalization clustering. The main contribution of this work is considering the vocalization spectrogram as an image (i.e., two-dimensional pattern) and representing it in terms of principal component analysis coefficients that feed a clustering approach. A performance study is carried out for each stage of the scheme. The methodology is tested to analyze three years of recordings from two wetlands in Panama to support ongoing efforts to estimate the manatee population. VC 2019 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license.
The Journal of the Acoustical Society of America
Acoustical Society of America