We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Acoustic Identification of Dolphin Whistle Types in Deep Waters of Arabian Sea Using Wavelet Threshold Denoising Approach.
- Authors
MAHANTY, Madan M.; CHEENANKANDY, Sanjana M.; LATHA, Ganesan; RAGURAMAN, Govindan; VENKATESAN, Ramasamy
- Abstract
In situ time series measurements of ocean ambient noise, have been made in deep waters of the Arabian Sea, using an autonomous passive acoustic monitoring system deployed as part of the Ocean Moored buoy network in the Northern Indian Ocean (OMNI) buoy mooring operated by the National Institute of Ocean Technology (NIOT), in Chennai during November 2018 to November 2019. The analysis of ambient noise records during the spring (April-June) showed the presence of dolphin whistles but contaminated by unwanted impulsive shackle noise. The frequency contours of the dolphin whistles occur in narrow band in the range 4-16 kHz. However, the unwanted impulsive shackle noise occurs in broad band with the noise level higher by -20 dB over the dolphin signals, and it reduces the quality of dolphin whistles. A wavelet based threshold denoising technique followed by a subtraction method is implemented. Reduction of unwanted shackle noise is effectively done and different dolphin whistle types are identified. This wavelet denoising approach is demonstrated for extraction of dolphin whistles in the presence of challenging impulsive shackle noise. Furthermore, this study should be useful for identifying other cetacean species when the signal of interest is interrupted by unwanted mechanical noise.
- Subjects
CHENNAI (India); BOTTLENOSE dolphin; DOLPHINS; SEAWATER; WHISTLES; TIME measurements; TIME series analysis; CETACEA
- Publication
Archives of Acoustics, 2023, Vol 48, Issue 1, p39
- ISSN
0137-5075
- Publication type
Article
- DOI
10.24425/aoa.2023.144264