Fish identification from videos captured in uncontrolled underwater environments

Faisal Shafait; Ajmal Mian; Mark Shortis; Bernard Ghanem; Phil F. Culverhouse; Duane Edgington; Danelle Cline; Mehdi Ravanbakhsh; James Seager; Euan S. Harvey

doi:10.1093/icesjms/fsw106

Fish identification from videos captured in uncontrolled underwater environments

Faisal Shafait, Ajmal Mian, Mark Shortis, Bernard Ghanem, Phil F. Culverhouse, Duane Edgington, Danelle Cline, Mehdi Ravanbakhsh, James Seager, Euan S. Harvey

Research output: Contribution to journal › Article › peer-review

Abstract

There is an urgent need for the development of sampling techniques which can provide accurate and precise count, size, and biomass data for fish. This information is essential to support the decision-making processes of fisheries and marine conservation managers and scientists. Digital video technology is rapidly improving, and it is now possible to record long periods of high resolution digital imagery cost effectively, making single or stereo-video systems one of the primary sampling tools. However, manual species identification, counting, and measuring of fish in stereo-video images is labour intensive and is the major disincentive against the uptake of this technology. Automating species identification using technologies developed by researchers in computer vision and machine learning would transform marine science. In this article, a new paradigm of image set classification is presented that can be used to achieve improved recognition rates for a number of fish species. State-of-the-art image set construction, modelling, and matching algorithms from computer vision literature are discussed with an analysis of their application for automatic fish species identification. It is demonstrated that these algorithms have the potential of solving the automatic fish species identification problem in underwater videos captured within unconstrained environments.

Original language	English (US)
Pages (from-to)	2737-2746
Number of pages	10
Journal	ICES Journal of Marine Science
Volume	73
Issue number	10
DOIs	https://doi.org/10.1093/icesjms/fsw106
State	Published - Nov 2016

Bibliographical note

KAUST Repository Item: Exported on 2021-07-08
Acknowledgements: We would like to thank the anonymous reviewers whose detailed and insightful comments have helped us in bringing this manuscript to its current form. In addition, the authors acknowledge support from the Australian Research Council Grant LP110201008, which provided the primary funding for this study in addition to UWA Research Collaboration Award (RCA) grant and King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research. A.M. was supported by the Australian Research Council Fellowship DP110102399.

Keywords

computer vision
fish classification
fish identification
image analysis
image sets
species recognition
COMPUTER VISION
SPECIES RECOGNITION
VISUAL TRACKING
CLASSIFICATION
SHAPE

ASJC Scopus subject areas

Ecology
Ecology, Evolution, Behavior and Systematics
Oceanography
Aquatic Science

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1093/icesjms/fsw106

Cite this

@article{0778eddf664d4f948dcc462e9cf0e513,

title = "Fish identification from videos captured in uncontrolled underwater environments",

abstract = "There is an urgent need for the development of sampling techniques which can provide accurate and precise count, size, and biomass data for fish. This information is essential to support the decision-making processes of fisheries and marine conservation managers and scientists. Digital video technology is rapidly improving, and it is now possible to record long periods of high resolution digital imagery cost effectively, making single or stereo-video systems one of the primary sampling tools. However, manual species identification, counting, and measuring of fish in stereo-video images is labour intensive and is the major disincentive against the uptake of this technology. Automating species identification using technologies developed by researchers in computer vision and machine learning would transform marine science. In this article, a new paradigm of image set classification is presented that can be used to achieve improved recognition rates for a number of fish species. State-of-the-art image set construction, modelling, and matching algorithms from computer vision literature are discussed with an analysis of their application for automatic fish species identification. It is demonstrated that these algorithms have the potential of solving the automatic fish species identification problem in underwater videos captured within unconstrained environments.",

keywords = "computer vision, fish classification, fish identification, image analysis, image sets, species recognition, COMPUTER VISION, SPECIES RECOGNITION, VISUAL TRACKING, CLASSIFICATION, SHAPE",

author = "Faisal Shafait and Ajmal Mian and Mark Shortis and Bernard Ghanem and Culverhouse, {Phil F.} and Duane Edgington and Danelle Cline and Mehdi Ravanbakhsh and James Seager and Harvey, {Euan S.}",

note = "KAUST Repository Item: Exported on 2021-07-08 Acknowledgements: We would like to thank the anonymous reviewers whose detailed and insightful comments have helped us in bringing this manuscript to its current form. In addition, the authors acknowledge support from the Australian Research Council Grant LP110201008, which provided the primary funding for this study in addition to UWA Research Collaboration Award (RCA) grant and King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research. A.M. was supported by the Australian Research Council Fellowship DP110102399.",

year = "2016",

month = nov,

doi = "10.1093/icesjms/fsw106",

language = "English (US)",

volume = "73",

pages = "2737--2746",

journal = "ICES Journal of Marine Science",

issn = "1095-9289",

publisher = "OXFORD UNIV PRESS",

number = "10",

}

TY - JOUR

T1 - Fish identification from videos captured in uncontrolled underwater environments

AU - Shafait, Faisal

AU - Mian, Ajmal

AU - Shortis, Mark

AU - Ghanem, Bernard

AU - Culverhouse, Phil F.

AU - Edgington, Duane

AU - Cline, Danelle

AU - Ravanbakhsh, Mehdi

AU - Seager, James

AU - Harvey, Euan S.

N1 - KAUST Repository Item: Exported on 2021-07-08 Acknowledgements: We would like to thank the anonymous reviewers whose detailed and insightful comments have helped us in bringing this manuscript to its current form. In addition, the authors acknowledge support from the Australian Research Council Grant LP110201008, which provided the primary funding for this study in addition to UWA Research Collaboration Award (RCA) grant and King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research. A.M. was supported by the Australian Research Council Fellowship DP110102399.

PY - 2016/11

Y1 - 2016/11

N2 - There is an urgent need for the development of sampling techniques which can provide accurate and precise count, size, and biomass data for fish. This information is essential to support the decision-making processes of fisheries and marine conservation managers and scientists. Digital video technology is rapidly improving, and it is now possible to record long periods of high resolution digital imagery cost effectively, making single or stereo-video systems one of the primary sampling tools. However, manual species identification, counting, and measuring of fish in stereo-video images is labour intensive and is the major disincentive against the uptake of this technology. Automating species identification using technologies developed by researchers in computer vision and machine learning would transform marine science. In this article, a new paradigm of image set classification is presented that can be used to achieve improved recognition rates for a number of fish species. State-of-the-art image set construction, modelling, and matching algorithms from computer vision literature are discussed with an analysis of their application for automatic fish species identification. It is demonstrated that these algorithms have the potential of solving the automatic fish species identification problem in underwater videos captured within unconstrained environments.

AB - There is an urgent need for the development of sampling techniques which can provide accurate and precise count, size, and biomass data for fish. This information is essential to support the decision-making processes of fisheries and marine conservation managers and scientists. Digital video technology is rapidly improving, and it is now possible to record long periods of high resolution digital imagery cost effectively, making single or stereo-video systems one of the primary sampling tools. However, manual species identification, counting, and measuring of fish in stereo-video images is labour intensive and is the major disincentive against the uptake of this technology. Automating species identification using technologies developed by researchers in computer vision and machine learning would transform marine science. In this article, a new paradigm of image set classification is presented that can be used to achieve improved recognition rates for a number of fish species. State-of-the-art image set construction, modelling, and matching algorithms from computer vision literature are discussed with an analysis of their application for automatic fish species identification. It is demonstrated that these algorithms have the potential of solving the automatic fish species identification problem in underwater videos captured within unconstrained environments.

KW - computer vision

KW - fish classification

KW - fish identification

KW - image analysis

KW - image sets

KW - species recognition

KW - COMPUTER VISION

KW - SPECIES RECOGNITION

KW - VISUAL TRACKING

KW - CLASSIFICATION

KW - SHAPE

UR - http://hdl.handle.net/10754/670058

UR - https://academic.oup.com/icesjms/article-lookup/doi/10.1093/icesjms/fsw106

U2 - 10.1093/icesjms/fsw106

DO - 10.1093/icesjms/fsw106

M3 - Article

SN - 1095-9289

VL - 73

SP - 2737

EP - 2746

JO - ICES Journal of Marine Science

JF - ICES Journal of Marine Science

IS - 10

ER -

Fish identification from videos captured in uncontrolled underwater environments

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this