Towards a Quantitative Analysis of Martian Terrains

This document investigates the use of sample based Bayesian segmentation algorithms on images of the Martian landscape, with regard to a quantitative understanding of statistical and systematic errors. The errors predicted from stochastic sampling are estimated using the conventional methods of error propagation and the Cramer-Rao bound and compared to the estimation errors for computed total quantities of textures. We show that systematic errors will rapidy grow to exceed the statistical errors for cases where the additive texture model does not discribe the data density distribution accurately. We investigate how these distributions might be better modelled using mixtures. We conclude that more work is still needed to apply such pattern recognition methods to scientific measurement.