Automatic annotation of coral reefs using deep learning

A. Mahmood, M. Bennamoun, S. An, F. Sohel, F. Boussaid, R. Hovey, G. Kendrick, R. B. Fisher

Research output: Chapter in Book/Conference paperConference paper

10 Citations (Scopus)

Abstract

Healthy coral reefs play a vital role in maintaining biodiversity in tropical marine ecosystems. Deep sea exploration and imaging have provided us with a great opportunity to look into the vast and complex marine ecosystems. Data acquisition from the coral reefs has facilitated the scientific investigation of these intricate ecosystems. Millions of digital images of the sea floor have been collected with the help of Remotely Operated Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). Automated technology to monitor the health of the oceans allows for transformational ecological outcomes by standardizing methods for detecting and identifying species. Manual annotation is a tediously repetitive and a time consuming task for marine experts. It takes 10-30 minutes for a marine expert to meticulously annotate a single image. This paper aims to automate the analysis of large available AUV imagery by developing advanced deep learning tools for rapid and large-scale automatic annotation of marine coral species. Such an automated technology would greatly benefit marine ecological studies in terms of cost, speed, accuracy and thus in better quantifying the level of environmental change marine ecosystems can tolerate. We propose a deep learning based classification method for coral reefs. We also report the application of the proposed technique towards the automatic annotation of unlabelled mosaics of the coral reef in the Abrolhos Islands, Western Australia. Our proposed method automatically quantifies the coral coverage in this region and detects a decreasing trend in coral population which is in line with conclusions by marine ecologists.

Original languageEnglish
Title of host publicationOCEANS 2016 MTS/IEEE Monterey, OCE 2016
EditorsJill Zande, Bill Kirkwood
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISBN (Electronic)9781509015375
DOIs
Publication statusPublished - 28 Nov 2016
Event2016 OCEANS MTS/IEEE Monterey, OCE 2016 - Monterey, United States
Duration: 19 Sep 201623 Sep 2016

Conference

Conference2016 OCEANS MTS/IEEE Monterey, OCE 2016
CountryUnited States
CityMonterey
Period19/09/1623/09/16

Fingerprint

coral reefs
annotations
Reefs
ecosystems
learning
Aquatic ecosystems
coral reef
marine ecosystem
underwater vehicles
coral
Autonomous underwater vehicles
autonomous underwater vehicle
biological diversity
Remotely operated vehicles
remotely operated vehicle
Biodiversity
digital image
imagery
data acquisition
Ecosystems

Cite this

Mahmood, A., Bennamoun, M., An, S., Sohel, F., Boussaid, F., Hovey, R., ... Fisher, R. B. (2016). Automatic annotation of coral reefs using deep learning. In J. Zande, & B. Kirkwood (Eds.), OCEANS 2016 MTS/IEEE Monterey, OCE 2016 [7761105] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/OCEANS.2016.7761105
Mahmood, A. ; Bennamoun, M. ; An, S. ; Sohel, F. ; Boussaid, F. ; Hovey, R. ; Kendrick, G. ; Fisher, R. B. / Automatic annotation of coral reefs using deep learning. OCEANS 2016 MTS/IEEE Monterey, OCE 2016. editor / Jill Zande ; Bill Kirkwood. IEEE, Institute of Electrical and Electronics Engineers, 2016.
@inproceedings{b59008f79660478c8f0ca907ed8377a2,
title = "Automatic annotation of coral reefs using deep learning",
abstract = "Healthy coral reefs play a vital role in maintaining biodiversity in tropical marine ecosystems. Deep sea exploration and imaging have provided us with a great opportunity to look into the vast and complex marine ecosystems. Data acquisition from the coral reefs has facilitated the scientific investigation of these intricate ecosystems. Millions of digital images of the sea floor have been collected with the help of Remotely Operated Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). Automated technology to monitor the health of the oceans allows for transformational ecological outcomes by standardizing methods for detecting and identifying species. Manual annotation is a tediously repetitive and a time consuming task for marine experts. It takes 10-30 minutes for a marine expert to meticulously annotate a single image. This paper aims to automate the analysis of large available AUV imagery by developing advanced deep learning tools for rapid and large-scale automatic annotation of marine coral species. Such an automated technology would greatly benefit marine ecological studies in terms of cost, speed, accuracy and thus in better quantifying the level of environmental change marine ecosystems can tolerate. We propose a deep learning based classification method for coral reefs. We also report the application of the proposed technique towards the automatic annotation of unlabelled mosaics of the coral reef in the Abrolhos Islands, Western Australia. Our proposed method automatically quantifies the coral coverage in this region and detects a decreasing trend in coral population which is in line with conclusions by marine ecologists.",
keywords = "Classification, Corals, Deep learning, Marine ecosystems, Marine images",
author = "A. Mahmood and M. Bennamoun and S. An and F. Sohel and F. Boussaid and R. Hovey and G. Kendrick and Fisher, {R. B.}",
year = "2016",
month = "11",
day = "28",
doi = "10.1109/OCEANS.2016.7761105",
language = "English",
editor = "Jill Zande and Kirkwood, {Bill }",
booktitle = "OCEANS 2016 MTS/IEEE Monterey, OCE 2016",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Mahmood, A, Bennamoun, M, An, S, Sohel, F, Boussaid, F, Hovey, R, Kendrick, G & Fisher, RB 2016, Automatic annotation of coral reefs using deep learning. in J Zande & B Kirkwood (eds), OCEANS 2016 MTS/IEEE Monterey, OCE 2016., 7761105, IEEE, Institute of Electrical and Electronics Engineers, 2016 OCEANS MTS/IEEE Monterey, OCE 2016, Monterey, United States, 19/09/16. https://doi.org/10.1109/OCEANS.2016.7761105

Automatic annotation of coral reefs using deep learning. / Mahmood, A.; Bennamoun, M.; An, S.; Sohel, F.; Boussaid, F.; Hovey, R.; Kendrick, G.; Fisher, R. B.

OCEANS 2016 MTS/IEEE Monterey, OCE 2016. ed. / Jill Zande; Bill Kirkwood. IEEE, Institute of Electrical and Electronics Engineers, 2016. 7761105.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Automatic annotation of coral reefs using deep learning

AU - Mahmood, A.

AU - Bennamoun, M.

AU - An, S.

AU - Sohel, F.

AU - Boussaid, F.

AU - Hovey, R.

AU - Kendrick, G.

AU - Fisher, R. B.

PY - 2016/11/28

Y1 - 2016/11/28

N2 - Healthy coral reefs play a vital role in maintaining biodiversity in tropical marine ecosystems. Deep sea exploration and imaging have provided us with a great opportunity to look into the vast and complex marine ecosystems. Data acquisition from the coral reefs has facilitated the scientific investigation of these intricate ecosystems. Millions of digital images of the sea floor have been collected with the help of Remotely Operated Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). Automated technology to monitor the health of the oceans allows for transformational ecological outcomes by standardizing methods for detecting and identifying species. Manual annotation is a tediously repetitive and a time consuming task for marine experts. It takes 10-30 minutes for a marine expert to meticulously annotate a single image. This paper aims to automate the analysis of large available AUV imagery by developing advanced deep learning tools for rapid and large-scale automatic annotation of marine coral species. Such an automated technology would greatly benefit marine ecological studies in terms of cost, speed, accuracy and thus in better quantifying the level of environmental change marine ecosystems can tolerate. We propose a deep learning based classification method for coral reefs. We also report the application of the proposed technique towards the automatic annotation of unlabelled mosaics of the coral reef in the Abrolhos Islands, Western Australia. Our proposed method automatically quantifies the coral coverage in this region and detects a decreasing trend in coral population which is in line with conclusions by marine ecologists.

AB - Healthy coral reefs play a vital role in maintaining biodiversity in tropical marine ecosystems. Deep sea exploration and imaging have provided us with a great opportunity to look into the vast and complex marine ecosystems. Data acquisition from the coral reefs has facilitated the scientific investigation of these intricate ecosystems. Millions of digital images of the sea floor have been collected with the help of Remotely Operated Vehicles (ROVs) and Autonomous Underwater Vehicles (AUVs). Automated technology to monitor the health of the oceans allows for transformational ecological outcomes by standardizing methods for detecting and identifying species. Manual annotation is a tediously repetitive and a time consuming task for marine experts. It takes 10-30 minutes for a marine expert to meticulously annotate a single image. This paper aims to automate the analysis of large available AUV imagery by developing advanced deep learning tools for rapid and large-scale automatic annotation of marine coral species. Such an automated technology would greatly benefit marine ecological studies in terms of cost, speed, accuracy and thus in better quantifying the level of environmental change marine ecosystems can tolerate. We propose a deep learning based classification method for coral reefs. We also report the application of the proposed technique towards the automatic annotation of unlabelled mosaics of the coral reef in the Abrolhos Islands, Western Australia. Our proposed method automatically quantifies the coral coverage in this region and detects a decreasing trend in coral population which is in line with conclusions by marine ecologists.

KW - Classification

KW - Corals

KW - Deep learning

KW - Marine ecosystems

KW - Marine images

UR - http://www.scopus.com/inward/record.url?scp=85006870201&partnerID=8YFLogxK

U2 - 10.1109/OCEANS.2016.7761105

DO - 10.1109/OCEANS.2016.7761105

M3 - Conference paper

BT - OCEANS 2016 MTS/IEEE Monterey, OCE 2016

A2 - Zande, Jill

A2 - Kirkwood, Bill

PB - IEEE, Institute of Electrical and Electronics Engineers

ER -

Mahmood A, Bennamoun M, An S, Sohel F, Boussaid F, Hovey R et al. Automatic annotation of coral reefs using deep learning. In Zande J, Kirkwood B, editors, OCEANS 2016 MTS/IEEE Monterey, OCE 2016. IEEE, Institute of Electrical and Electronics Engineers. 2016. 7761105 https://doi.org/10.1109/OCEANS.2016.7761105