Probabilistic graphical techniques for automated ice-bottom tracking and comparison between state-of-the-art solutions

Abstract

We present improvements to existing frameworks for automated extraction of ice interfaces applied to two-dimensional and three-dimensional radar echograms of polar ice sheets. These improvements consist of novel image pre-processing steps and empirically-derived cost functions that allow for the integration of further domain-specific knowledge into the models employed. Along with an explanation of our modifications, we demonstrate the results obtained by our proposed models and algorithms, such as a 43% decrease in mean tracking error in the case of three-dimensional imagery. We also present the results obtained by several state-of-the-art ice-interface tracking solutions, and compare all automated results with manually-corrected ground-truth data. Furthermore, we perform a self-assessment of tracking results by analyzing the differences found between the automatically extracted ice-layers in cases where two separate radar measurements have been made at the same location.