Seafloor Classification based on an AUV Based Sub-bottom Acoustic Probe Data for Mn-crust survey

TL;DR

This paper presents a method for automatic seafloor classification using sub-bottom acoustic data collected by an AUV, achieving around 70% accuracy in deep-sea Mn-crust survey regions.

Contribution

It introduces a novel approach combining autoencoders and SVMs for classifying seafloor types from acoustic reflections, reducing reliance on visual inspection.

Findings

Achieved approximately 70% classification accuracy.

Demonstrated feasibility of automated seafloor classification from acoustic data.

Validated method in deep-sea Mn-crust survey conditions.

Abstract

The possibility of automatically classifying high frequency sub-bottom acoustic reflections collected from an Autonomous Underwater Robot is investigated in this paper. In field surveys of Cobalt-rich Manganese Crusts (Mn-crusts), existing methods relies on visual confirmation of seafloor from images and thickness measurements using the sub-bottom probe. Using these visual classification results as ground truth, an autoencoder is trained to extract latent features from bundled acoustic reflections. A Support Vector Machine classifier is then trained to classify the latent space to idetify seafloor classes. Results from data collected from seafloor at 1500m deep regions of Mn-crust showed an accuracy of about 70%.