Please use this identifier to cite or link to this item:
|Title:||Automatic template matching for classification of dolphin vocalizations|
|Authors:||Rui, G. |
|Keywords:||Dynamic time warping|
|Citation:||Rui, G.,Chitre, M.,Ong, S.H.,Taylor, E. (2008). Automatic template matching for classification of dolphin vocalizations. OCEANS'08 MTS/IEEE Kobe-Techno-Ocean'08 - Voyage toward the Future, OTO'08 : -. ScholarBank@NUS Repository. https://doi.org/10.1109/OCEANSKOBE.2008.4530981|
|Abstract:||Whistle classification is a key step in many studies of dolphin vocalizations. Automatic whistles tracing algorithms have been developed but tracing errors such as breaks and outliers are usually unavoidable. Local variations of whistle contours occur even in whistles of the same type. In this paper, we describe a modified dynamic time warping (DTW) algorithm for dynamic non-linear matching. It exhibits a good performance in matching against a template whistle. The modifications to the basic DTW algorithm provide improved tolerance to noise and breaks in tracing. Together with automatic de-noising, this template matching is used to classify vocalizations of Indo-Pacific dolphins (Sousa chinensis). We believe this method can be applied for large scale analysis of whistles for species recognition, dolphin training and other dolphin studies. ©2008 IEEE.|
|Source Title:||OCEANS'08 MTS/IEEE Kobe-Techno-Ocean'08 - Voyage toward the Future, OTO'08|
|Appears in Collections:||Staff Publications|
Show full item record
Files in This Item:
There are no files associated with this item.
checked on Jan 10, 2019
checked on Jan 12, 2019
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.