mirage   mirage   mirage

Acoustic classification of multiple simultaneous bird species: A multi-instance multi-label approach

DSpace/Manakin Repository

Show simple item record

dc.creator Briggs, Forrest
dc.creator Lakshminarayanan, Balaji
dc.creator Neal, Lawrence
dc.creator Fern, Xiaoli Z.
dc.creator Raich, Raviv
dc.creator Hadley, Sarah J. K.
dc.creator Hadley, Adam S.
dc.creator Betts, Matthew G.
dc.date.accessioned 2012-11-16T22:06:43Z
dc.date.available 2013-10-15T22:18:26Z
dc.date.issued 2012-06
dc.identifier.citation Briggs, F., Lakshminarayanan, B., Neal, L., Fern, X. Z., Raich, R., Hadley, S. J. K., . . . . (2012). Acoustic classification of multiple simultaneous bird species: A multi-instance multi-label approach. The Journal of the Acoustical Society of America, 131(6), 4640-4650. doi: 10.1121/1.4707424 en_US
dc.identifier.uri http://hdl.handle.net/1957/35097
dc.description This is the publisher’s final pdf. The published article is copyrighted by the Acoustical Society of America and can be found at: http://asadl.org/jasa/. en_US
dc.description.abstract Although field-collected recordings typically contain multiple simultaneously vocalizing birds of different species, acoustic species classification in this setting has received little study so far. This work formulates the problem of classifying the set of species present in an audio recording using the multi-instance multi-label (MIML) framework for machine learning, and proposes a MIML bag generator for audio, i.e., an algorithm which transforms an input audio signal into a bag-of-instances representation suitable for use with MIML classifiers. The proposed representation uses a 2D time-frequency segmentation of the audio signal, which can separate bird sounds that overlap in time. Experiments using audio data containing 13 species collected with unattended omnidirectional microphones in the H. J. Andrews Experimental Forest demonstrate that the proposed methods achieve high accuracy (96.1% true positives/negatives). Automated detection of bird species occurrence using MIML has many potential applications, particularly in long-term monitoring of remote sites, species distribution modeling, and conservation planning. en_US
dc.description.sponsorship This work was partially funded by the Ecosystems Informatics IGERT program via NSF Grant No. DGE 0333257, NSF-CDI Grant No. 0941748 to M.G.B., NSF Grant No. 1055113 to X.Z.F., and the College of Engineering, Oregon State University. We conducted this research at H. J. Andrews Experimental Forest, which is funded by the US Forest Service, Pacific Northwest Research Station. en_US
dc.language.iso en_US en_US
dc.publisher Acoustical Society of America en_US
dc.relation.ispartofseries Journal of the Acoustical Society of America en_US
dc.relation.ispartofseries Vol. 131 no. 6 en_US
dc.title Acoustic classification of multiple simultaneous bird species: A multi-instance multi-label approach en_US
dc.type Article en_US
dc.description.peerreview yes en_US
dc.identifier.doi 10.1121/1.4707424
dc.description.embargopolicy Repository Administrators en


This item appears in the following Collection(s)

Show simple item record

Search ScholarsArchive@OSU


Advanced Search

Browse

My Account

Statistics