Minke whale detection in underwater imagery using classification CNNs
A predictable aggregation of dwarf minke whales occurs annually in the Australian offshore waters of the northern Great Barrier Reef in June-July, which has been the subject of a long-term photo-identification study. Researchers from the Minke Whale Project (MWP) at James Cook University collect lar...
Published in: | Global Oceans 2020: Singapore – U.S. Gulf Coast |
---|---|
Main Authors: | , , , , , , , , , , |
Format: | Conference Object |
Language: | unknown |
Published: |
Institute of Electrical and Electronics Engineers
2020
|
Subjects: | |
Online Access: | https://researchonline.jcu.edu.au/67937/1/viewpaper_OCEANS2020.pdf |
id |
ftjamescook:oai:researchonline.jcu.edu.au:67937 |
---|---|
record_format |
openpolar |
spelling |
ftjamescook:oai:researchonline.jcu.edu.au:67937 2024-02-11T10:05:51+01:00 Minke whale detection in underwater imagery using classification CNNs Konovalov, Dmitry A. Swinhoe, Natalie Efremova, Dina B. Birtles, R. Alastair Kusetic, Martha Adams, Kent Hillcoat, Suzanne Curnock, Matthew I. Williams, Genevieve Sobtzick, Susan Sheaves, Marcus 2020 application/pdf https://researchonline.jcu.edu.au/67937/1/viewpaper_OCEANS2020.pdf unknown Institute of Electrical and Electronics Engineers https://doi.org/10.1109/IEEECONF38699.2020.9389164 https://researchonline.jcu.edu.au/67937/ https://researchonline.jcu.edu.au/67937/1/viewpaper_OCEANS2020.pdf Konovalov, Dmitry A., Swinhoe, Natalie, Efremova, Dina B., Birtles, R. Alastair, Kusetic, Martha, Adams, Kent, Hillcoat, Suzanne, Curnock, Matthew I., Williams, Genevieve, Sobtzick, Susan, and Sheaves, Marcus (2020) Minke whale detection in underwater imagery using classification CNNs. In: Proceedings of Global Oceans 2020. From: Global Oceans 2020: Singapore – U.S. Gulf Coast, 5-30 October 2020, Biloxi, MS, USA. restricted Conference Item PeerReviewed 2020 ftjamescook https://doi.org/10.1109/IEEECONF38699.2020.9389164 2024-01-22T23:48:11Z A predictable aggregation of dwarf minke whales occurs annually in the Australian offshore waters of the northern Great Barrier Reef in June-July, which has been the subject of a long-term photo-identification study. Researchers from the Minke Whale Project (MWP) at James Cook University collect large volumes of underwater digital imagery each season (e.g. 1.8TB in 2018), much of which is contributed by citizen scientists. Manual processing and analysis of this quantity of data had become infeasible, and Convolutional Neural Networks (CNNs) offered a potential solution. Our study sought to design and train a CNN that could detect whales from video footage in complex near-surface underwater surroundings containing multiple peo- ple, boats, research and recreational gear. We modified known classification CNNs to localize whales in video frames and digital still images. The required high detection accuracy was achieved by discovering an effective negative-labeling training technique. This resulted in a less than 1% false-positive detection rate and below 0.1% false-negative rate. The final operation-version CNN- pipeline processed all videos (with the interval of 10 frames) in approximately four days (running on two GPUs) delivering 1.95 million sorted images. Conference Object minke whale James Cook University, Australia: ResearchOnline@JCU Global Oceans 2020: Singapore – U.S. Gulf Coast 1 9 |
institution |
Open Polar |
collection |
James Cook University, Australia: ResearchOnline@JCU |
op_collection_id |
ftjamescook |
language |
unknown |
description |
A predictable aggregation of dwarf minke whales occurs annually in the Australian offshore waters of the northern Great Barrier Reef in June-July, which has been the subject of a long-term photo-identification study. Researchers from the Minke Whale Project (MWP) at James Cook University collect large volumes of underwater digital imagery each season (e.g. 1.8TB in 2018), much of which is contributed by citizen scientists. Manual processing and analysis of this quantity of data had become infeasible, and Convolutional Neural Networks (CNNs) offered a potential solution. Our study sought to design and train a CNN that could detect whales from video footage in complex near-surface underwater surroundings containing multiple peo- ple, boats, research and recreational gear. We modified known classification CNNs to localize whales in video frames and digital still images. The required high detection accuracy was achieved by discovering an effective negative-labeling training technique. This resulted in a less than 1% false-positive detection rate and below 0.1% false-negative rate. The final operation-version CNN- pipeline processed all videos (with the interval of 10 frames) in approximately four days (running on two GPUs) delivering 1.95 million sorted images. |
format |
Conference Object |
author |
Konovalov, Dmitry A. Swinhoe, Natalie Efremova, Dina B. Birtles, R. Alastair Kusetic, Martha Adams, Kent Hillcoat, Suzanne Curnock, Matthew I. Williams, Genevieve Sobtzick, Susan Sheaves, Marcus |
spellingShingle |
Konovalov, Dmitry A. Swinhoe, Natalie Efremova, Dina B. Birtles, R. Alastair Kusetic, Martha Adams, Kent Hillcoat, Suzanne Curnock, Matthew I. Williams, Genevieve Sobtzick, Susan Sheaves, Marcus Minke whale detection in underwater imagery using classification CNNs |
author_facet |
Konovalov, Dmitry A. Swinhoe, Natalie Efremova, Dina B. Birtles, R. Alastair Kusetic, Martha Adams, Kent Hillcoat, Suzanne Curnock, Matthew I. Williams, Genevieve Sobtzick, Susan Sheaves, Marcus |
author_sort |
Konovalov, Dmitry A. |
title |
Minke whale detection in underwater imagery using classification CNNs |
title_short |
Minke whale detection in underwater imagery using classification CNNs |
title_full |
Minke whale detection in underwater imagery using classification CNNs |
title_fullStr |
Minke whale detection in underwater imagery using classification CNNs |
title_full_unstemmed |
Minke whale detection in underwater imagery using classification CNNs |
title_sort |
minke whale detection in underwater imagery using classification cnns |
publisher |
Institute of Electrical and Electronics Engineers |
publishDate |
2020 |
url |
https://researchonline.jcu.edu.au/67937/1/viewpaper_OCEANS2020.pdf |
genre |
minke whale |
genre_facet |
minke whale |
op_relation |
https://doi.org/10.1109/IEEECONF38699.2020.9389164 https://researchonline.jcu.edu.au/67937/ https://researchonline.jcu.edu.au/67937/1/viewpaper_OCEANS2020.pdf Konovalov, Dmitry A., Swinhoe, Natalie, Efremova, Dina B., Birtles, R. Alastair, Kusetic, Martha, Adams, Kent, Hillcoat, Suzanne, Curnock, Matthew I., Williams, Genevieve, Sobtzick, Susan, and Sheaves, Marcus (2020) Minke whale detection in underwater imagery using classification CNNs. In: Proceedings of Global Oceans 2020. From: Global Oceans 2020: Singapore – U.S. Gulf Coast, 5-30 October 2020, Biloxi, MS, USA. |
op_rights |
restricted |
op_doi |
https://doi.org/10.1109/IEEECONF38699.2020.9389164 |
container_title |
Global Oceans 2020: Singapore – U.S. Gulf Coast |
container_start_page |
1 |
op_container_end_page |
9 |
_version_ |
1790603062719545344 |