Research Lines: Marine Bioacoustic Signal Processing
Passive Acoustic Monitoring (PAM) of underwater sounds can be used to detect the presence of cetacean species and to guarantee that anthropogenic noise meets acceptable limits as the EU Marine Strategy Framework Directive (MSFD) determines. The MSFD aims to achieve what it is defined as Good Environmental State (GES) of our oceans and seas. Signal processing can play a key role automating the analysis of large PAM recordings and helping to extract information about the presence, abundance and habits of different species.
Different machine learning techniques (including deep learning) are applied to detect, track, and classify underwater calls. We also create new graphical representation techniques in order to help in the analysis of very long temporal series such as those obtained from 1/3 octave continuous underwater noise.
We have done numerous PAM campaigns in different locations. Some of the most representatives are: Cabrera Archipelago Maritime National Park (Balearic Islands), Columbretes Islands Natural Park (Castellon), and Cape of Sant Antoni (Denia). For these marine protected areas, specific graph analysis tools were designed to reveal seasonal structures of human made noises, while at the same time, allowing to visualize cetacean presence. We also design fast algorithms to detect cetacean calls extracting their contours and classifying them according to different criteria.
Authored published scientific papers:
- R. Miralles, G. Lara-Martínez, L. Carrión-Garcia, and M. Bou-Cabo. Assessment of Arrow-of-Time Metrics for the Characterization of Underwater Explosions. Sensors, 21:1–15, 2021.
Download: [pdf] [ps.gz] [ps] [HTML] - R. Verling, E. And Miralles, M. Bou-Cabo, G. Lara-Martínez, M. Garagouni, J.M. Brignon, and T. O'Higgins. Application of a risk-based approach to continuous underwater noise at local and subregional scales for the Marine Strategy Framework Directive. Marine Policy, 134:1–13, 2021.
Download: [pdf] [ps.gz] [ps] [HTML] - Ramon Miralles-Ricos, Guillermo Lara, Ramón Miralles, Manuel Bou-Cabo, Jose Antonio Esteban, and Victor Espinosa. New Insights into the Design and Application of a Passive Acoustic Monitoring System for the Assessment of the Good Environmental Status in Spanish Marine Waters. Sensors, 2020.
Download: [pdf] [ps.gz] [ps] [HTML] - Ramón Miralles, Guillermo Lara, Jorge Gosalbez, Ignacio Bosch, and Antonio León. Improved visualization of large temporal series for the evaluation of good environmental status. Applied Acoustics, 148:55–61, Elsevier BV, may 2019.
Download: [pdf] [ps.gz] [ps] [HTML] - A. Carrión, G. Lara, R. Miralles, J. Gosálbez, and I. Bosch. Characterizing the mechanisms of sound production in odontocetes: A signal modality approach. International Conference on Digital Signal Processing, DSP, 2017-August, 2017.
Download: [pdf] - R. Miralles, G. Lara, A. Carrión, J. Gosalbez, and I. Bosch. On the detection of impulsive and tonal events in passive acoustics monitoring. International Conference on Digital Signal Processing, DSP, 2017-August, 2017.
Download: [pdf] - R. Miralles, G. Lara, A. Carrion, and J.A. Esteban. Automatic Detection and Classification of Beluga Whale Vocalizations. Advances in Applied Acoustics (AIAA), 2(2):61–70, 2013.
Download: [pdf] - R. Miralles, G. Lara, A. Carrion, and J.A. Esteban. SAMARUC a Programmable system for Passive acoustic monitoring of cetaceans. WAVES, 5(1):69–79, 2013.
Download: [pdf] - R. Miralles, G. Lara, J.A. Esteban, and A. Rodriguez. The pulsed to tonal strength parameter and its importance in characterizing and classifying Beluga whale sounds. J. Acoust. Soc. Am., 131(3):2173–2179, March 2012. Matlab code available
Download: [pdf] [ps.gz] [ps] [HTML]