Automatic species recognition using echolocation clicks from odontocetes

摘要

The classification of different odontocetes using écholocation clicks plays a significant role in tracking and detecting animals for research and protection purposes. Echolocation clicks were detected by an automatic method based on the Teager-Kaiser Energy Operator (TKEO). Then, these clicks were represented by their FFT magnitude spectrum. To reduce the influence of high similar clicks among species, a joint decision of the k nearest neighbors method with reject option (JDKNNRO) was proposed to recognize the species. Experiments were conducted on a dataset collected from odontocetes inhabiting Xiamen Bay, and that from the 3rd International Workshop on the Detection and Classification of Marine Mammals. Both datasets contain three species. The average accuracies of species identification were 92.38% and 99.30% respectively, which indicated the effectiveness of JDKNNRO.