Reconocimiento automático de especies utilizando procesamiento digital de imágenes acústicas

• Published in: 2014 IEEE Biennial Congress of Argentina (ARGENCON) pp. 25 - 30
• Main Authors: Villar, Sebastian A., Madirolas, Adrian, Mosquera, Maximiliano, Cabreira, Ariel, Rossi, Silvano R., Acosta, Gerardo G.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-06-2014
• Subjects: Image color analysis; Image edge detection; Organisms; Software; Software algorithms; Sonar; Surface treatment
• DOI: 10.1109/ARGENCON.2014.6868467

Accurate identification of marine organisms and their numerical abundance calculation using echo detection techniques remains a challenge for researchers. Different approaches using hydroacoustic techniques have been applied, alone or combined, to study marine and fresh water environments. An alternative approach to this problem is to use echoic statistics with confident geometric descriptors obtained from the echo-records processing. To accomplish this task it is necessary to have a development platform that allows reading echo-records from a particular echo-sounder, to detect aggregations or schools and then to calculate the various descriptors that will be use for species identification, in an automatic way. This article describes the application of digital processing algorithms of echo-records for automatic recognition of the ocean seabed, surface and marine organisms. These algorithms are implemented within the EcoPampa software, which is the first Argentinean open source system for the identification of marine species.