A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images
Abstract Fish counts and species information can be obtained from images taken within trawls, which enables trawl surveys to operate without extracting fish from their habitat, yields distribution data at fine scale for better interpretation of acoustic results, and can detect fish that are not reta...
Published in: | ICES Journal of Marine Science |
---|---|
Main Authors: | , , , |
Other Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Oxford University Press (OUP)
2021
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1093/icesjms/fsab227 https://academic.oup.com/icesjms/article-pdf/78/10/3780/41772702/fsab227.pdf |
id |
croxfordunivpr:10.1093/icesjms/fsab227 |
---|---|
record_format |
openpolar |
spelling |
croxfordunivpr:10.1093/icesjms/fsab227 2024-05-19T07:46:19+00:00 A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images Allken, Vaneeda Rosen, Shale Handegard, Nils Olav Malde, Ketil Demer, David Research Council of Norway Norwegian Ministry of Trade, Industry and Fisheries 2021 http://dx.doi.org/10.1093/icesjms/fsab227 https://academic.oup.com/icesjms/article-pdf/78/10/3780/41772702/fsab227.pdf en eng Oxford University Press (OUP) https://creativecommons.org/licenses/by/4.0/ ICES Journal of Marine Science volume 78, issue 10, page 3780-3792 ISSN 1054-3139 1095-9289 journal-article 2021 croxfordunivpr https://doi.org/10.1093/icesjms/fsab227 2024-05-02T09:31:04Z Abstract Fish counts and species information can be obtained from images taken within trawls, which enables trawl surveys to operate without extracting fish from their habitat, yields distribution data at fine scale for better interpretation of acoustic results, and can detect fish that are not retained in the catch due to mesh selection. To automate the process of image-based fish detection and identification, we trained a deep learning algorithm (RetinaNet) on images collected from the trawl-mounted Deep Vision camera system. In this study, we focused on the detection of blue whiting, Atlantic herring, Atlantic mackerel, and mesopelagic fishes from images collected in the Norwegian sea. To address the need for large amounts of annotated data to train these models, we used a combination of real and synthetic images, and obtained a mean average precision of 0.845 on a test set of 918 images. Regression models were used to compare predicted fish counts, which were derived from RetinaNet classification of fish in the individual image frames, with catch data collected at 20 trawl stations. We have automatically detected and counted fish from individual images, related these counts to the trawl catches, and discussed how to use this in regular trawl surveys. Article in Journal/Newspaper Norwegian Sea Oxford University Press ICES Journal of Marine Science 78 10 3780 3792 |
institution |
Open Polar |
collection |
Oxford University Press |
op_collection_id |
croxfordunivpr |
language |
English |
description |
Abstract Fish counts and species information can be obtained from images taken within trawls, which enables trawl surveys to operate without extracting fish from their habitat, yields distribution data at fine scale for better interpretation of acoustic results, and can detect fish that are not retained in the catch due to mesh selection. To automate the process of image-based fish detection and identification, we trained a deep learning algorithm (RetinaNet) on images collected from the trawl-mounted Deep Vision camera system. In this study, we focused on the detection of blue whiting, Atlantic herring, Atlantic mackerel, and mesopelagic fishes from images collected in the Norwegian sea. To address the need for large amounts of annotated data to train these models, we used a combination of real and synthetic images, and obtained a mean average precision of 0.845 on a test set of 918 images. Regression models were used to compare predicted fish counts, which were derived from RetinaNet classification of fish in the individual image frames, with catch data collected at 20 trawl stations. We have automatically detected and counted fish from individual images, related these counts to the trawl catches, and discussed how to use this in regular trawl surveys. |
author2 |
Demer, David Research Council of Norway Norwegian Ministry of Trade, Industry and Fisheries |
format |
Article in Journal/Newspaper |
author |
Allken, Vaneeda Rosen, Shale Handegard, Nils Olav Malde, Ketil |
spellingShingle |
Allken, Vaneeda Rosen, Shale Handegard, Nils Olav Malde, Ketil A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
author_facet |
Allken, Vaneeda Rosen, Shale Handegard, Nils Olav Malde, Ketil |
author_sort |
Allken, Vaneeda |
title |
A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
title_short |
A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
title_full |
A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
title_fullStr |
A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
title_full_unstemmed |
A deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
title_sort |
deep learning-based method to identify and count pelagic and mesopelagic fishes from trawl camera images |
publisher |
Oxford University Press (OUP) |
publishDate |
2021 |
url |
http://dx.doi.org/10.1093/icesjms/fsab227 https://academic.oup.com/icesjms/article-pdf/78/10/3780/41772702/fsab227.pdf |
genre |
Norwegian Sea |
genre_facet |
Norwegian Sea |
op_source |
ICES Journal of Marine Science volume 78, issue 10, page 3780-3792 ISSN 1054-3139 1095-9289 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.1093/icesjms/fsab227 |
container_title |
ICES Journal of Marine Science |
container_volume |
78 |
container_issue |
10 |
container_start_page |
3780 |
op_container_end_page |
3792 |
_version_ |
1799486468379377664 |