Unsupervised seabed characterization with Bayesian modeling of SAS imagery

Abstract

Seabed characterization has utility for numerous applications that seek to explore and interact with the seafloor, ranging from costal habitat monitoring and sub-bottom profiling to man-made object detection. In the work presented here, we characterize the seabed based on the texture patterns within SAS images constructed from high-frequency side-scan sonar. Features are measured from the SAS images (e.g., lacunarity, an established texture feature coding method, and a rotationally invariant histogram of oriented gradients). Based on these SAS image features, we perform unsupervised clustering with a hierarchical Bayesian model, which creates categories of seabed textures. Our clustering model is a new variant of the hierarchical Dirichlet process that is both adaptive to changes in the seabed and processes batches of SAS imagery in an online fashion. This allows the model to learn new seabed types as they are encountered and provides usable clustering results after each batch is processed, rather than having to wait for all the data to be collected before being processed. The model’s performance of seabed characterization by SAS image texture is demonstrated in the overall range and internal consistency of textures specific to each learned cluster. [Work supported by the Office of Naval Research.]