Towards Automated Classification of Seabed Substrates in Underwater Video

Matthew Pugh; Bernie Tiddeman; Hannah Mary Dee; Phil Hughes

doi:10.1109/CVAUI.2014.18

Towards Automated Classification of Seabed Substrates in Underwater Video

Matthew Pugh, Bernie Tiddeman, Hannah Mary Dee, Phil Hughes

Cyfrifiadureg

Friends of Cardigan Bay

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Trafodion Cynhadledd (Nid-Cyfnodolyn fathau)

187 Wedi eu Llwytho i Lawr (Pure)

Crynodeb

In this work, we present a system for the automated classiffication of seabed substrates in underwater video. Classiffication of seabed substrates traditionally requires manual analysis by a marine biologist, according to an established classiffication system. Accurate, consistent and robust classiffication is difficult in underwater video due to varying lighting conditions, turbidity and method of original recording. We have developed a system that uses ground truth data from marine biologists to train and test per-frame classiffiers. In this paper we present preliminary results of this using various feature representations (histograms, Gabor wavelets) and classiffiers (SVC, kNN) on both full-frame and patched-based analysis, achieving up to 93% accuracy.

Iaith wreiddiol	Saesneg
Teitl	Computer Vision for Analysis of Underwater lmagery Workshop
Is-deitl	International Conference on Pattern Recognition
Man cyhoeddi	Stockholm
Cyhoeddwr	IEEE Press
Tudalennau	9-16
Nifer y tudalennau	8
ISBN (Argraffiad)	978-1-4799-6709-4
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.1109/CVAUI.2014.18
Statws	Cyhoeddwyd - 24 Awst 2014
Digwyddiad	22nd International Conference on Pattern Recognition - Stockholm Waterfront, Stockholm, Sweden Hyd: 24 Awst 2014 → 28 Awst 2014

Cynhadledd

Cynhadledd	22nd International Conference on Pattern Recognition
Gwlad/Tiriogaeth	Sweden
Dinas	Stockholm
Cyfnod	24 Awst 2014 → 28 Awst 2014

NDC y CU

Mae’r allbwn hwn yn cyfrannu at y Nod(au) Datblygu Cynaliadwy canlynol

Mynediad at Ddogfen

10.1109/CVAUI.2014.18

Towards automated classification of seabed substrates in underwater videoLlawysgrif awdur wedi’i dderbyn, 2.33 MB

Cysylltiad parhaol

Dolen barhaus

Ffeiliau eraill a chysylltiadau

http://hdl.handle.net/2160/30549

Dyfynnu hyn

@inproceedings{1e8ceead07a3427fbe956d3940490374,

title = "Towards Automated Classification of Seabed Substrates in Underwater Video",

abstract = "In this work, we present a system for the automated classiffication of seabed substrates in underwater video. Classiffication of seabed substrates traditionally requires manual analysis by a marine biologist, according to an established classiffication system. Accurate, consistent and robust classiffication is difficult in underwater video due to varying lighting conditions, turbidity and method of original recording. We have developed a system that uses ground truth data from marine biologists to train and test per-frame classiffiers. In this paper we present preliminary results of this using various feature representations (histograms, Gabor wavelets) and classiffiers (SVC, kNN) on both full-frame and patched-based analysis, achieving up to 93% accuracy.",

keywords = "underwater video analysis, texture, Gabor filters, substrate classification, machine learning",

author = "Matthew Pugh and Bernie Tiddeman and Dee, {Hannah Mary} and Phil Hughes",

year = "2014",

month = aug,

day = "24",

doi = "10.1109/CVAUI.2014.18",

language = "English",

isbn = "978-1-4799-6709-4",

pages = "9--16",

booktitle = "Computer Vision for Analysis of Underwater lmagery Workshop",

publisher = "IEEE Press",

address = "United States of America",

note = "22nd International Conference on Pattern Recognition ; Conference date: 24-08-2014 Through 28-08-2014",

}

Pugh, M, Tiddeman, B , Dee, HM & Hughes, P 2014, Towards Automated Classification of Seabed Substrates in Underwater Video. yn Computer Vision for Analysis of Underwater lmagery Workshop: International Conference on Pattern Recognition. IEEE Press, Stockholm, tt. 9-16, 22nd International Conference on Pattern Recognition, Stockholm, Sweden, 24 Awst 2014. https://doi.org/10.1109/CVAUI.2014.18

Towards Automated Classification of Seabed Substrates in Underwater Video. / Pugh, Matthew; Tiddeman, Bernie ; Dee, Hannah Mary et al.
Computer Vision for Analysis of Underwater lmagery Workshop: International Conference on Pattern Recognition. Stockholm: IEEE Press, 2014. t. 9-16.

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Trafodion Cynhadledd (Nid-Cyfnodolyn fathau)

TY - GEN

T1 - Towards Automated Classification of Seabed Substrates in Underwater Video

AU - Pugh, Matthew

AU - Tiddeman, Bernie

AU - Dee, Hannah Mary

AU - Hughes, Phil

PY - 2014/8/24

Y1 - 2014/8/24

N2 - In this work, we present a system for the automated classiffication of seabed substrates in underwater video. Classiffication of seabed substrates traditionally requires manual analysis by a marine biologist, according to an established classiffication system. Accurate, consistent and robust classiffication is difficult in underwater video due to varying lighting conditions, turbidity and method of original recording. We have developed a system that uses ground truth data from marine biologists to train and test per-frame classiffiers. In this paper we present preliminary results of this using various feature representations (histograms, Gabor wavelets) and classiffiers (SVC, kNN) on both full-frame and patched-based analysis, achieving up to 93% accuracy.

AB - In this work, we present a system for the automated classiffication of seabed substrates in underwater video. Classiffication of seabed substrates traditionally requires manual analysis by a marine biologist, according to an established classiffication system. Accurate, consistent and robust classiffication is difficult in underwater video due to varying lighting conditions, turbidity and method of original recording. We have developed a system that uses ground truth data from marine biologists to train and test per-frame classiffiers. In this paper we present preliminary results of this using various feature representations (histograms, Gabor wavelets) and classiffiers (SVC, kNN) on both full-frame and patched-based analysis, achieving up to 93% accuracy.

KW - underwater video analysis

KW - texture

KW - Gabor filters

KW - substrate classification

KW - machine learning

UR - http://hdl.handle.net/2160/30549

U2 - 10.1109/CVAUI.2014.18

DO - 10.1109/CVAUI.2014.18

M3 - Conference Proceeding (Non-Journal item)

SN - 978-1-4799-6709-4

SP - 9

EP - 16

BT - Computer Vision for Analysis of Underwater lmagery Workshop

PB - IEEE Press

CY - Stockholm

T2 - 22nd International Conference on Pattern Recognition

Y2 - 24 August 2014 through 28 August 2014

ER -

Towards Automated Classification of Seabed Substrates in Underwater Video

Crynodeb

Cynhadledd

NDC y CU

Mynediad at Ddogfen

Cysylltiad parhaol

Ffeiliau eraill a chysylltiadau

Ôl bys

Dyfynnu hyn