Assessment of ZooImage as a tool for the classification of zooplankton
ZooImage, image analysis software, was evaluated to determine its ability to differentiate between zooplankton groups in preserved zooplankton samples collected in Prince William Sound, Alaska. A training set of 53 categories was established to train the software for automatic recognition. Using the...
Published in: | Journal of Plankton Research |
---|---|
Main Authors: | , |
Format: | Text |
Language: | English |
Published: |
Oxford University Press
2008
|
Subjects: | |
Online Access: | http://plankt.oxfordjournals.org/cgi/content/short/fbn092v1 https://doi.org/10.1093/plankt/fbn092 |
id |
fthighwire:oai:open-archive.highwire.org:plankt:fbn092v1 |
---|---|
record_format |
openpolar |
spelling |
fthighwire:oai:open-archive.highwire.org:plankt:fbn092v1 2023-05-15T18:48:56+02:00 Assessment of ZooImage as a tool for the classification of zooplankton Bell, Jenefer Hopcroft, Russell R. 2008-09-15 04:59:48.0 text/html http://plankt.oxfordjournals.org/cgi/content/short/fbn092v1 https://doi.org/10.1093/plankt/fbn092 en eng Oxford University Press http://plankt.oxfordjournals.org/cgi/content/short/fbn092v1 http://dx.doi.org/10.1093/plankt/fbn092 Copyright (C) 2008, Oxford University Press Article TEXT 2008 fthighwire https://doi.org/10.1093/plankt/fbn092 2016-11-16T18:35:41Z ZooImage, image analysis software, was evaluated to determine its ability to differentiate between zooplankton groups in preserved zooplankton samples collected in Prince William Sound, Alaska. A training set of 53 categories was established to train the software for automatic recognition. Using the Random forest algorithm, ZooImage identified particles in the training set with less than 13% error. Despite reasonable results with the training set, however, ZooImage was less effective when this training set was used to identify particles from field-collected zooplankton samples. When all particles were examined, ZooImage had an accuracy of 81.7 % but this dropped to 63.3% when discard particles (e.g. marine snow, fibers) were removed from total particles. Because copepods were the numerically dominant organisms in most samples, they were examined separately and were correctly identified 67.8% of the time. Further investigation suggested size was effective in determining identifications; medium size copepods (e.g., Pseudocalanus sp., Acartia sp.) were accurately identified 73.3% of the time. ZooImage can provide a coarse level of taxonomic classification and we anticipate continued improvement to this software should further enhance automatic identification of preserved zooplankton samples. Text Alaska Copepods HighWire Press (Stanford University) Journal of Plankton Research 30 12 1351 1367 |
institution |
Open Polar |
collection |
HighWire Press (Stanford University) |
op_collection_id |
fthighwire |
language |
English |
topic |
Article |
spellingShingle |
Article Bell, Jenefer Hopcroft, Russell R. Assessment of ZooImage as a tool for the classification of zooplankton |
topic_facet |
Article |
description |
ZooImage, image analysis software, was evaluated to determine its ability to differentiate between zooplankton groups in preserved zooplankton samples collected in Prince William Sound, Alaska. A training set of 53 categories was established to train the software for automatic recognition. Using the Random forest algorithm, ZooImage identified particles in the training set with less than 13% error. Despite reasonable results with the training set, however, ZooImage was less effective when this training set was used to identify particles from field-collected zooplankton samples. When all particles were examined, ZooImage had an accuracy of 81.7 % but this dropped to 63.3% when discard particles (e.g. marine snow, fibers) were removed from total particles. Because copepods were the numerically dominant organisms in most samples, they were examined separately and were correctly identified 67.8% of the time. Further investigation suggested size was effective in determining identifications; medium size copepods (e.g., Pseudocalanus sp., Acartia sp.) were accurately identified 73.3% of the time. ZooImage can provide a coarse level of taxonomic classification and we anticipate continued improvement to this software should further enhance automatic identification of preserved zooplankton samples. |
format |
Text |
author |
Bell, Jenefer Hopcroft, Russell R. |
author_facet |
Bell, Jenefer Hopcroft, Russell R. |
author_sort |
Bell, Jenefer |
title |
Assessment of ZooImage as a tool for the classification of zooplankton |
title_short |
Assessment of ZooImage as a tool for the classification of zooplankton |
title_full |
Assessment of ZooImage as a tool for the classification of zooplankton |
title_fullStr |
Assessment of ZooImage as a tool for the classification of zooplankton |
title_full_unstemmed |
Assessment of ZooImage as a tool for the classification of zooplankton |
title_sort |
assessment of zooimage as a tool for the classification of zooplankton |
publisher |
Oxford University Press |
publishDate |
2008 |
url |
http://plankt.oxfordjournals.org/cgi/content/short/fbn092v1 https://doi.org/10.1093/plankt/fbn092 |
genre |
Alaska Copepods |
genre_facet |
Alaska Copepods |
op_relation |
http://plankt.oxfordjournals.org/cgi/content/short/fbn092v1 http://dx.doi.org/10.1093/plankt/fbn092 |
op_rights |
Copyright (C) 2008, Oxford University Press |
op_doi |
https://doi.org/10.1093/plankt/fbn092 |
container_title |
Journal of Plankton Research |
container_volume |
30 |
container_issue |
12 |
container_start_page |
1351 |
op_container_end_page |
1367 |
_version_ |
1766242311975469056 |