Testing the performances of automated identification of bat echolocation calls : A request for prudence
Echolocating bats are surveyed and studied acoustically with bat detectors routinely and worldwide, yet identification of species from calls often remains ambiguous or impossible due to intraspecific call variation and/or interspecific overlap in call design. To overcome such difficulties and to red...
Published in: | Ecological Indicators |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Elsevier
2017
|
Subjects: | |
Online Access: | https://lup.lub.lu.se/record/2db131c4-85b3-4e1a-8d52-35344299da2a https://doi.org/10.1016/j.ecolind.2017.03.023 |
id |
ftulundlup:oai:lup.lub.lu.se:2db131c4-85b3-4e1a-8d52-35344299da2a |
---|---|
record_format |
openpolar |
spelling |
ftulundlup:oai:lup.lub.lu.se:2db131c4-85b3-4e1a-8d52-35344299da2a 2024-05-12T08:01:37+00:00 Testing the performances of automated identification of bat echolocation calls : A request for prudence Rydell, Jens Nyman, Stefan Eklöf, Johan Jones, Gareth A. Russo, Danilo 2017-07-01 https://lup.lub.lu.se/record/2db131c4-85b3-4e1a-8d52-35344299da2a https://doi.org/10.1016/j.ecolind.2017.03.023 eng eng Elsevier https://lup.lub.lu.se/record/2db131c4-85b3-4e1a-8d52-35344299da2a http://dx.doi.org/10.1016/j.ecolind.2017.03.023 wos:000406435900043 scopus:85016419173 Ecological Indicators; 78, pp 416-420 (2017) ISSN: 1470-160X Zoology Biosonar Methodology Software Species identification Ultrasound contributiontojournal/article info:eu-repo/semantics/article text 2017 ftulundlup https://doi.org/10.1016/j.ecolind.2017.03.023 2024-04-17T14:03:32Z Echolocating bats are surveyed and studied acoustically with bat detectors routinely and worldwide, yet identification of species from calls often remains ambiguous or impossible due to intraspecific call variation and/or interspecific overlap in call design. To overcome such difficulties and to reduce workload, automated classifiers of echolocation calls have become popular, but their performance has not been tested sufficiently in the field. We examined the absolute performance of two commercially available programs (SonoChiro and Kaleidoscope) and one freeware package (BatClassify). We recorded noise from rain and calls of seven common bat species with Pettersson real-time full spectrum detectors in Sweden. The programs could always (100%) distinguish rain from bat calls, usually (68–100%) identify bats to group (Nyctalus/Vespertilio/Eptesicus, Pipistrellus, Myotis, Plecotus, Barbastella) and usually (83–99%) recognize typical calls of some species whose echolocation pulses are structurally distinct (Pipistrellus pygmaeus, Barbastella barbastellus). Species with less characteristic echolocation calls were not identified reliably, including Vespertilio murinus (16–26%), Myotis spp. (4–93%) and Plecotus auritus (0–89%). All programs showed major although different shortcomings and the often poor performance raising serious concerns about the use of automated classifiers for identification to species level in research and surveys. We highlight the importance of validating output from automated classifiers, and restricting their use to specific situations where identification can be made with high confidence. For comparison we also present the result of a manual identification test on a random subset of the files used to test the programs. It showed a higher classification success but performances were still low for more problematic taxa. Article in Journal/Newspaper Barbastella barbastellus Vespertilio murinus Lund University Publications (LUP) Ecological Indicators 78 416 420 |
institution |
Open Polar |
collection |
Lund University Publications (LUP) |
op_collection_id |
ftulundlup |
language |
English |
topic |
Zoology Biosonar Methodology Software Species identification Ultrasound |
spellingShingle |
Zoology Biosonar Methodology Software Species identification Ultrasound Rydell, Jens Nyman, Stefan Eklöf, Johan Jones, Gareth A. Russo, Danilo Testing the performances of automated identification of bat echolocation calls : A request for prudence |
topic_facet |
Zoology Biosonar Methodology Software Species identification Ultrasound |
description |
Echolocating bats are surveyed and studied acoustically with bat detectors routinely and worldwide, yet identification of species from calls often remains ambiguous or impossible due to intraspecific call variation and/or interspecific overlap in call design. To overcome such difficulties and to reduce workload, automated classifiers of echolocation calls have become popular, but their performance has not been tested sufficiently in the field. We examined the absolute performance of two commercially available programs (SonoChiro and Kaleidoscope) and one freeware package (BatClassify). We recorded noise from rain and calls of seven common bat species with Pettersson real-time full spectrum detectors in Sweden. The programs could always (100%) distinguish rain from bat calls, usually (68–100%) identify bats to group (Nyctalus/Vespertilio/Eptesicus, Pipistrellus, Myotis, Plecotus, Barbastella) and usually (83–99%) recognize typical calls of some species whose echolocation pulses are structurally distinct (Pipistrellus pygmaeus, Barbastella barbastellus). Species with less characteristic echolocation calls were not identified reliably, including Vespertilio murinus (16–26%), Myotis spp. (4–93%) and Plecotus auritus (0–89%). All programs showed major although different shortcomings and the often poor performance raising serious concerns about the use of automated classifiers for identification to species level in research and surveys. We highlight the importance of validating output from automated classifiers, and restricting their use to specific situations where identification can be made with high confidence. For comparison we also present the result of a manual identification test on a random subset of the files used to test the programs. It showed a higher classification success but performances were still low for more problematic taxa. |
format |
Article in Journal/Newspaper |
author |
Rydell, Jens Nyman, Stefan Eklöf, Johan Jones, Gareth A. Russo, Danilo |
author_facet |
Rydell, Jens Nyman, Stefan Eklöf, Johan Jones, Gareth A. Russo, Danilo |
author_sort |
Rydell, Jens |
title |
Testing the performances of automated identification of bat echolocation calls : A request for prudence |
title_short |
Testing the performances of automated identification of bat echolocation calls : A request for prudence |
title_full |
Testing the performances of automated identification of bat echolocation calls : A request for prudence |
title_fullStr |
Testing the performances of automated identification of bat echolocation calls : A request for prudence |
title_full_unstemmed |
Testing the performances of automated identification of bat echolocation calls : A request for prudence |
title_sort |
testing the performances of automated identification of bat echolocation calls : a request for prudence |
publisher |
Elsevier |
publishDate |
2017 |
url |
https://lup.lub.lu.se/record/2db131c4-85b3-4e1a-8d52-35344299da2a https://doi.org/10.1016/j.ecolind.2017.03.023 |
genre |
Barbastella barbastellus Vespertilio murinus |
genre_facet |
Barbastella barbastellus Vespertilio murinus |
op_source |
Ecological Indicators; 78, pp 416-420 (2017) ISSN: 1470-160X |
op_relation |
https://lup.lub.lu.se/record/2db131c4-85b3-4e1a-8d52-35344299da2a http://dx.doi.org/10.1016/j.ecolind.2017.03.023 wos:000406435900043 scopus:85016419173 |
op_doi |
https://doi.org/10.1016/j.ecolind.2017.03.023 |
container_title |
Ecological Indicators |
container_volume |
78 |
container_start_page |
416 |
op_container_end_page |
420 |
_version_ |
1798843725246365696 |