Ground-Level Classification of a Coral Reef Using a Hyperspectral Camera

Abstract

Especially in the remote sensing context, thematic classification is a desired product for coral reef surveys. This study presents a novel statistical-based image classification approach, namely Partial Least Square Discriminant Analysis (PLS-DA), capable of doing so. Three classification models were built and implemented for the images while the fourth was a combination of spectra from all three images together. The classification was optimised by using pre-processing transformations (PPTs) and post-classification low-pass filtering. Despite the fact that the images were acquired under different conditions and quality, the best classification model was achieved by combining spectral training samples from three images (accuracy 0.63 for all classes). PPTs improved the classification accuracy by 5%-15% and post-classification treatments further increased the final accuracy by 10%-20%. The fourth classification model was the most accurate one, suggesting that combining spectra from differ conditions improves thematic classification. Despite some limitations, available aerial sensors already provide an opportunity to implement the described classification and mark the next investigation step. Nonetheless, the findings of this study are relevant both to the field of remote sensing in general and to the niche of coral reef spectroscopy.

Towards improved estimates of sea-ice algal biomass: experimental assessment of hyperspectral imaging cameras for under-ice studies

“…The main advantage of the proposed application of HI is the ability to capture light at high spectral and spatial resolutions (in this case mm scales) in a non-invasive manner with great potential for studies aimed at investigating ice-algal spatial distribution, variability and environmental controls (Campbell and others, 2015; Lange and others, 2016; Lund-Hansen and others, 2016). Studies employing hyperspectral imaging for different types of underwater biomes have been performed on mycrophytobenthos (Chennu and others, 2013) and coral reef biota (Caras and Karnieli, 2015).…”

An Under-Ice Hyperspectral and RGB Imaging System to Capture Fine-Scale Biophysical Properties of Sea Ice

“…In theory, hyperspectral resolution data has the potential to resolve beyond pure biomass estimates towards more sophisticated biological traits such as ice algal photophysiology [91,97,98], species composition [97,99,100], pigment detection [101][102][103], and feature classification and mapping [35,76,104]. An interesting field is also being explored in the retrieval of primary production estimates from spectral data in combination with in-vitro photosynthetic parameters for ice algae [7,105] or with PAM fluorometry for microphytobenthic communities [106].…”

Spectral Reflectance of Palauan Reef-Building Coral with Different Symbionts in Response to Elevated Temperature

“…The relationship between integrated reflectance and symbiont concentration may prove useful to assess or monitor reef condition and the potential for bleaching. Such algorithms may be implemented using hyperspectral sensors from airborne platforms like the Portable Remote Imaging Spectrometer (PRISM) [1,38], proposed future satellite sensors, and through the use of underwater hyperspectral imagers [32,33,47,48]. However, the implementation of the algorithms will be complicated by the effects of the intervening water column, sea-surface, and atmosphere, as well as the spectral and spatial resolution of the sensor itself [24,35].…”