Development of Artificial Intelligence Methods for Hyperspectral Image Analysis in Coral Reef Science Applications

Graduate Research School For Candidates Prospective Candidates Available Projects Development of Artificial Intelligence Methods for Hyperspectral Image Analysis in Coral Reef Science Applications

Development of Artificial Intelligence Methods for Hyperspectral Image Analysis in Coral Reef Science Applications

Title of Project

Development of Artificial Intelligence Methods for Hyperspectral Image Analysis in Coral Reef Science Applications

Advisor/s

Jonathan Kok

College or Research Centre

AIMS@JCU

Summary of Project

The aim of this research project is to investigate the use of advanced artificial intelligence (AI) methods for hyperspectral image analysis in coral reef science applications (such as, bleaching prediction, benthic community properties, etc.). Current hyperspectral analysis methods rely on satellite and low spatial resolution (e.g., 30 to 250 metres per pixel) hyperspectral imagery, which require significant pixel un-mixing and complex atmospheric correction. Instead, the scope of this project is to use close range, millimetre resolution hyperspectral imagery from live coral tanks using AIMS hyperspectral and SeaSim infrastructure. The datasets will be used for developing a spectral library of high-quality end members as well as for temporal coral bleaching studies to identify stress indicators of bleaching before bleaching actually occurs. Due to the complexity and large amount of information inherent in hyperspectral datasets, traditional statistical analysis methods are not computationally feasible, hence AI methods (such as machine learning, neural network, deep learning, evolutionary algorithms, etc.) will be explored and evaluated. Methods developed will be further validated with field experiments. The output of this project will be valuable in bridging the gap of applying high quality hyperspectral data in coral reef science.

Key Words

hyperspectral imaging; hyperspectral analysis; coral reef; bleaching; neural network; deep learning; machine learning; artificial intelligence

Would suit an applicant who

Background in AI methods, spectroscopy and hyperspectral systems. Above average programming skills, i.e., able to code and modify AI algorithms. Familiarity with MATLAB would be ideal, but other programming languages are acceptable as well.

Updated: 1 month ago