Automated tracking of dolphin whistles using Gaussian Mixture Probability Hypothesis Density (GM-PHD) filters
Automated tracking of dolphin whistles using Gaussian Mixture Probability Hypothesis Density (GM-PHD) filters
This work considers automated Multi Target Tracking (MTT) of odontocete whistle contours. An adaptation of Gaussian Mixture Probability Hypothesis Density (GM-PHD)filter is described and applied to the acoustic recordings from six odontocete species. From the raw data, spectral peaks are first identified and then GM-PHD filter is used to simultaneously track the whistles’ frequency contours. Overall over 9000 whistles are tracked with a precision of 85% and recall of 71.8%. The proposed filter is shown to track whistles precisely (with mean deviation of 104 Hz, about one frequency bin, from the annotated whistle path) and 80% coverage. The filter is computationally efficient, suitable for real-time implementation, and is widely applicable to different odontocete species.
1-47
Gruden, Pina
2d951d33-121b-4bce-9cf4-daa3935e2fb4
White, Paul R.
2dd2477b-5aa9-42e2-9d19-0806d994eaba
Gruden, Pina
2d951d33-121b-4bce-9cf4-daa3935e2fb4
White, Paul R.
2dd2477b-5aa9-42e2-9d19-0806d994eaba
Gruden, Pina and White, Paul R.
(2016)
Automated tracking of dolphin whistles using Gaussian Mixture Probability Hypothesis Density (GM-PHD) filters.
Journal of the Acoustical Society of America, .
(doi:10.1121/1.4962980).
Abstract
This work considers automated Multi Target Tracking (MTT) of odontocete whistle contours. An adaptation of Gaussian Mixture Probability Hypothesis Density (GM-PHD)filter is described and applied to the acoustic recordings from six odontocete species. From the raw data, spectral peaks are first identified and then GM-PHD filter is used to simultaneously track the whistles’ frequency contours. Overall over 9000 whistles are tracked with a precision of 85% and recall of 71.8%. The proposed filter is shown to track whistles precisely (with mean deviation of 104 Hz, about one frequency bin, from the annotated whistle path) and 80% coverage. The filter is computationally efficient, suitable for real-time implementation, and is widely applicable to different odontocete species.
Text
JASA-00512_R1.pdf
- Accepted Manuscript
More information
Accepted/In Press date: 5 September 2016
e-pub ahead of print date: 23 September 2016
Organisations:
Inst. Sound & Vibration Research
Identifiers
Local EPrints ID: 400139
URI: http://eprints.soton.ac.uk/id/eprint/400139
ISSN: 0001-4966
PURE UUID: 8ce50790-5895-43f2-9171-ef08c046d00e
Catalogue record
Date deposited: 07 Sep 2016 10:42
Last modified: 15 Mar 2024 05:52
Export record
Altmetrics
Contributors
Author:
Pina Gruden
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
