Automated Identification Method for Detection and Classification of Neotropical Bats

Bats are indicators for ecosystem health, and therefore the determination of bat activity and species abundance provides essential information for biodiversity research and conservation monitoring. In this study, we propose a computational method for the automated identification of bat species based on echolocation calls. Our approach aims on performing in complex soundscapes such as those present in tropical ecosystems. We implement a statistical model-based Voice Activity Detector combined with an ensemble of supervised learning classifiers using Random Forest algorithm. We conducted a performance test using a library of Neotropical bat echolocation calls comprising 36 species. Results show that Random Forest classifiers had the most accurate performance among five supervised learning classifiers. Our identification method is not restricted on using only the rather stereotypic search phase calls but can also perform on more variable calls from other echolocation phases.