In vitro coral bleach observation using an RGB-IR camera

Pracharat Sa-Ngadsup, Yasushi Kiyoki, Chawan Koopipat

研究成果: Conference contribution

抄録

As global warming is worsening recently, coral bleaching is also happening more frequently because corals are sensitive to temperature change. Consequently, the need to study and monitor coral reefs is rapidly increasing in order to better understand the impact of environmental change on ocean ecology. There are two major approaches to monitoring coral reef. First, remote sensing, by using satellite imagery and/or aerial photography coral can be studied in large quantity in a very short period of time. The downside of this approach is that images obtained by satellite normally do not have enough spatial resolution and its images are not able to record very deep underwater objects. The second approach is using a diver to perform a detailed investigation of coral reef, which gives a lot more detail, but this approach is very time-consuming and expensive. Our aim is to use a small, commercially available camera as a tool to automatically monitor coral in real-time with the least possible human intervention. The main purpose of this research is to demonstrate that a consumer camera, with a little modification, can be used to detect coral bleaching. The modification was done so that the camera can record images in near-infrared range, which is important for coral detection. In this study, 5 coral species (Porites sp., Turbinaria sp., Acropora sp., Goniopora sp.and Pavona sp.), representing common local species, were induced to bleach by exposure to high water temperature for 24 hours. Images were taken periodically and 6 indices were constructed from those images. The indices which were IRR (Infrared-red), IRG (Infrared-Green), IRB (Infrared-Blue) and 3 variations of NDVI (Normal Difference Vegetation Index) were used to analyse the health of corals. The result shows that out of 6 indices being test the IRB index and NDVI (Blue) are the most promising pair for coral bleaching detection task.

本文言語English
ホスト出版物のタイトルInformation Modelling and Knowledge Bases XXX
編集者Tatiana Endrjukaite, Hannu Jaakkola, Alexander Dudko, Yasushi Kiyoki, Bernhard Thalheim, Naofumi Yoshida
出版社IOS Press
ページ457-468
ページ数12
ISBN(電子版)9781614999324
DOI
出版ステータスPublished - 2019

出版物シリーズ

名前Frontiers in Artificial Intelligence and Applications
312
ISSN(印刷版)0922-6389

ASJC Scopus subject areas

  • 人工知能

フィンガープリント

「In vitro coral bleach observation using an RGB-IR camera」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル