Sparse representation for classification of dolphin whistles by type

Automatic fish sounds classification

The work presented in this paper focuses on the use of acoustic systems for passive acoustic monitoring of ocean vitality for fish populations. Specifically, it focuses on the use of acoustic systems for passive acoustic monitoring of ocean vitality for fish populations. To this end, various indicators can be used to monitor marine areas such as both the geographical and temporal evolution of fish populations. A discriminative model is built using supervised machine learning (random-forest and support-vector machines). Each acquisition is represented in a feature space, in which the patterns belonging to different semantic classes are as separable as possible. The set of features proposed for describing the acquisitions come from an extensive state of the art in various domains in which classification of acoustic signals is performed, including speech, music, and environmental acoustics. Furthermore, this study proposes to extract features from three representations of the data (time, frequency, and cepstral domains). The proposed classification scheme is tested on real fish sounds recorded on several areas, and achieves 96.9% correct classification compared to 72.5% when using reference state of the art features as descriptors. The classification scheme is also validated on continuous underwater recordings, thereby illustrating that it can be used to both detect and classify fish sounds in operational scenarios.

Regional differences in the whistles of Australasian humpback dolphins (genus Sousa)

Characteristics of whistles may be used to study differentiation in dolphins to complement morphological and genetic studies. The whistles of four populations of Chinese humpback dolphins (Sousa chinensis chinensis (Osbeck, 1765)), one population of Taiwanese humpback dolphins (Sousa chinensis taiwanensis Wang, Yang, and Hung, 2015), and one population of Australian humpback dolphins (Sousa sahulensis Jefferson and Rosenbaum, 2014) were compared to determine if differences in whistles support current views of population structure and regional and species differentiation in the genus Sousa Gray, 1866. Acoustic features were extracted from whistles captured by broadband recording systems. Permutational MANOVAs were conducted to test for differences between populations, regions, and species. Random forest trees were also used to classify similar whistles. A significant amount of variation in acoustic features was explained by population (pseudo F[5,2742] = 191.66, p < 0.001), regional (pseudo F[3,2741] = 280.62, p < 0.001), and species (pseudo F[1,999] = 3.7, p < 0.05) differences in humpback dolphin whistles. Random forest trees correctly classified whistles into populations from 40% to 67%, regions from 51% to 80%, and species from 74% to 80%. Differences in whistles were consistent with the current ideas of population-, regional-, and species-level differences within the genus Sousa, based on morphological and genetic data, as well as geographic distance and barriers to movement.