Please use this identifier to cite or link to this item:
Title: Computer-based classification of dolphin whistles
Authors: GAO RUI
Keywords: dolphin, whistle, classification
Issue Date: 11-Feb-2011
Citation: GAO RUI (2011-02-11). Computer-based classification of dolphin whistles. ScholarBank@NUS Repository.
Abstract: Classification of dolphin whistles is essential for dolphin recognition and studies. We divide it into three steps: features selection, similarity between features, and classification method. One common feature vector describing whistles is a sequence of sample points along the whistle contour in spectrograms. Compared with human perception, the difficulty matching whistles is to deal with the local speed variation. Another problem is the local frequency variation. We present dynamic time warping (DTW) for nonlinearly matching. We also propose a compact feature vector - segments. A systematic procedure of classification is introduced: the feature vector records the curvatures, dynamic matching is established by a smoothing method. Comparison experiments show that our method outperforms the conventional methods, with more agreement with human classification. Together with the work of whistle extraction, systematic software is to be constructed to help dolphin researchers for automatic whistle detection and classifications from underwater recordings.
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
MasterThesis_GaoRui_2011.pdf5.34 MBAdobe PDF



Page view(s)

checked on Apr 19, 2019


checked on Apr 19, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.