Automated CNN Based Coral Reef Classification Using Image Augmentation and Deep Learning

Authors

  • Sumit Sharan School of Engineering and Physics, The University of the South Pacific, Laucala campus, Suva, Fiji
  • H. Harsh School of Engineering and Physics, The University of the South Pacific, Laucala campus, Suva, Fiji
  • Stuart Kininmonth The University of Queensland, Heron Island Research Station, QLD Australia
  • Utkal Mehta School of Engineering and Physics, The University of the South Pacific, Laucala campus, Suva, Fiji

Abstract

A critical issue faced by the marine scientist is to classify underwater images describing coral benthic cover. Typically, scientists take underwater imagery using high-resolution cameras and further analysis on these corals and marine species is done on land (preferably a laboratory) and by visual inspection. However, the analysis is time consuming, since the first step, which is the classification of corals, is an intensive activity by taxonomic experts. This traditional manual classification method is difficult to automate or quicken which is problematic given the high volume of images. In this work, the fundamental analysis is discussed by using available techniques such as deep learning (DL) and Convolutional Neural Network (CNN). It is required to find an easier, efficient and faster way to automate the classification of corals. This task is complicated, since most of the common coral
species look similar to one another. For reasons of structural diversity, it is easier to differentiate other forms of marine life such as fish and stingrays. This paper is based on the difficult but important Scleractinian (Stony) corals only. A technique recommended is investigated further at structural level such as branching corals. Verification result proves that the training and testing data are almost similar, thus the proposed technique is capable to learn and predict correctly.

Keywords: Coral classification, CNN, Automation, Deep learning, RGB approach.

Cite As

S. Sharan, Harsh, S. Kininmonth, U. Mehta, "Automated CNN Based Coral Reef Classification Using
Image Augmentation and Deep Learning", Engineering Intelligent Systems, vol. 29 no. 4, pp. 253-261,
2021.






Downloads

Published

2021-07-01

Issue

Vol. 29 No. 4 (2021): International Journal of Engineering Intelligent Systems

Section

General Submission

License

The submission of a paper implies that, if accepted for publication, it will not be published elsewhere in the same form, in any language, without the prior consent of the publisher. Before publication, authors are requested to assign copyright to CRL Publishing Ltd. This allows CRL to sanction photocopying, and to authorize the reprinting of issues or volumes according to demand. Authors' traditional rights will not be jeopardized by assigning Copyright in this way, as they retain the right to reuse the material following publication, and to veto third-party publication.