Towards Automated Classification of Seabed Substrates in Underwater Video

In this work, we present a system for the automated classification of seabed substrates in underwater video. Classification of seabed substrates traditionally requires manual analysis by a marine biologist, according to an established classification system. Accurate, consistent and robust classification is difficult in underwater video due to varying lighting conditions, turbidity and method of original recording. We have developed a system that uses ground truth data from marine biologists to train and test per-frame classifiers. In this paper we present preliminary results of this using various feature representations (histograms, Gabor wavelets) and classifiers (SVC, kNN) on both full-frame and patchedbased analysis, achieving up to 93% accuracy ...