Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision
With growing anthropogenic pressure on deep-sea ecosystems, large quantities of data are needed to understand their ecology, monitor changes over time and inform conservation managers. Current methods of image analysis are too slow to meet these requirements. Recently, computer vision has become mor...
Published in: | Ecological Informatics |
---|---|
Main Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Elsevier
2022
|
Subjects: | |
Online Access: | http://hdl.handle.net/10026.1/19761 https://doi.org/10.1016/j.ecoinf.2022.101786 |
id |
ftunivplympearl:oai:pearl.plymouth.ac.uk:10026.1/19761 |
---|---|
record_format |
openpolar |
spelling |
ftunivplympearl:oai:pearl.plymouth.ac.uk:10026.1/19761 2024-06-09T07:48:25+00:00 Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision Piechaud, N Howell, KL 2022-11 101786-101786 application/pdf http://hdl.handle.net/10026.1/19761 https://doi.org/10.1016/j.ecoinf.2022.101786 en eng Elsevier ISSN:1878-0512 E-ISSN:1878-0512 1878-0512 101786 http://hdl.handle.net/10026.1/19761 doi:10.1016/j.ecoinf.2022.101786 2022-10-28 Not known Benthic ecology Computer vision Xenophyophores Quantitative ecology Mapping Automated image analysis Marine conservation journal-article Article 2022 ftunivplympearl https://doi.org/10.1016/j.ecoinf.2022.101786 2024-05-14T23:48:21Z With growing anthropogenic pressure on deep-sea ecosystems, large quantities of data are needed to understand their ecology, monitor changes over time and inform conservation managers. Current methods of image analysis are too slow to meet these requirements. Recently, computer vision has become more accessible to biologists, and could help address this challenge. In this study we demonstrate a method by which non-specialists can train a YOLOV4 Convolutional Neural Network (CNN) able to count and measure a single class of objects. We apply CV to the extraction of quantitative data on the density and population size structure of the xenophyophore Syringammina fragilissima, from more than 58,000 images taken by an AUV 1200 m deep in the North-East Atlantic. The workflow developed used open-source tools, cloud-base hardware, and only required a level of experience with CV commonly found among ecologists. The CNN performed well, achieving a recall of 0.84 and precision of 0.91. Individual counts per image and size measurements resulting from model predictions were highly correlated (0.96 and 0.92, respectively) with manually collected data. The analysis could be completed in less than 10 days thus bringing novel insights into the population size structure and fine scale distribution of this Vulnerable Marine Ecosystem. It showed S. fragilissima distribution is patchy. The average density is 2.5 ind.m−2 but can vary from up to 45 ind.m−2 only a few tens of meter away from areas where it is almost absent. The average size is 5.5 cm and the largest individuals (>15 cm) tend to be in areas of low density. This study demonstrates how researchers could take advantage of CV to quickly and efficiently generate large quantitative datasets data on benthic ecosystems extent and distribution. This, coupled with the large sampling capacity of AUVs could bypass the bottleneck of image analysis and greatly facilitate future deep-ocean exploration and monitoring. It also illustrates the future potential of these new technologies ... Article in Journal/Newspaper North East Atlantic PEARL (Plymouth Electronic Archiv & ResearchLibrary, Plymouth University) Ecological Informatics 71 101786 |
institution |
Open Polar |
collection |
PEARL (Plymouth Electronic Archiv & ResearchLibrary, Plymouth University) |
op_collection_id |
ftunivplympearl |
language |
English |
topic |
Benthic ecology Computer vision Xenophyophores Quantitative ecology Mapping Automated image analysis Marine conservation |
spellingShingle |
Benthic ecology Computer vision Xenophyophores Quantitative ecology Mapping Automated image analysis Marine conservation Piechaud, N Howell, KL Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
topic_facet |
Benthic ecology Computer vision Xenophyophores Quantitative ecology Mapping Automated image analysis Marine conservation |
description |
With growing anthropogenic pressure on deep-sea ecosystems, large quantities of data are needed to understand their ecology, monitor changes over time and inform conservation managers. Current methods of image analysis are too slow to meet these requirements. Recently, computer vision has become more accessible to biologists, and could help address this challenge. In this study we demonstrate a method by which non-specialists can train a YOLOV4 Convolutional Neural Network (CNN) able to count and measure a single class of objects. We apply CV to the extraction of quantitative data on the density and population size structure of the xenophyophore Syringammina fragilissima, from more than 58,000 images taken by an AUV 1200 m deep in the North-East Atlantic. The workflow developed used open-source tools, cloud-base hardware, and only required a level of experience with CV commonly found among ecologists. The CNN performed well, achieving a recall of 0.84 and precision of 0.91. Individual counts per image and size measurements resulting from model predictions were highly correlated (0.96 and 0.92, respectively) with manually collected data. The analysis could be completed in less than 10 days thus bringing novel insights into the population size structure and fine scale distribution of this Vulnerable Marine Ecosystem. It showed S. fragilissima distribution is patchy. The average density is 2.5 ind.m−2 but can vary from up to 45 ind.m−2 only a few tens of meter away from areas where it is almost absent. The average size is 5.5 cm and the largest individuals (>15 cm) tend to be in areas of low density. This study demonstrates how researchers could take advantage of CV to quickly and efficiently generate large quantitative datasets data on benthic ecosystems extent and distribution. This, coupled with the large sampling capacity of AUVs could bypass the bottleneck of image analysis and greatly facilitate future deep-ocean exploration and monitoring. It also illustrates the future potential of these new technologies ... |
format |
Article in Journal/Newspaper |
author |
Piechaud, N Howell, KL |
author_facet |
Piechaud, N Howell, KL |
author_sort |
Piechaud, N |
title |
Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
title_short |
Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
title_full |
Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
title_fullStr |
Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
title_full_unstemmed |
Fast and accurate mapping of fine scale abundance of a VME in the deep sea with computer vision |
title_sort |
fast and accurate mapping of fine scale abundance of a vme in the deep sea with computer vision |
publisher |
Elsevier |
publishDate |
2022 |
url |
http://hdl.handle.net/10026.1/19761 https://doi.org/10.1016/j.ecoinf.2022.101786 |
genre |
North East Atlantic |
genre_facet |
North East Atlantic |
op_relation |
ISSN:1878-0512 E-ISSN:1878-0512 1878-0512 101786 http://hdl.handle.net/10026.1/19761 doi:10.1016/j.ecoinf.2022.101786 |
op_rights |
2022-10-28 Not known |
op_doi |
https://doi.org/10.1016/j.ecoinf.2022.101786 |
container_title |
Ecological Informatics |
container_volume |
71 |
container_start_page |
101786 |
_version_ |
1801380137129738240 |