Speaker
Description
It is well known that the muontomographic inverse problem is considerably underdetermined for several reasons, and therefore various types of regularization methods are required for its solution, which significantly influence the estimation bias and the covariance matrix of the estimated parameters. Following an overview of applicable regularization methods, the presentation examines several important properties of the Bayesian-type method linked to a priori information for the inversion of subsurface muographic measurements. It deals in detail with the relationship between the prior model and spatial resolution, the tuning of the prior density distribution to the near and far zones of the detector and analyzes the spatial distribution of bias and the regularization effect appearing in the variance of the results. The presentation also will discuss the symmetries characteristic of muographic mapping, which necessitate regularization.
The results of the analysis are presented by examining tomographic reconstruction based on field measurement data collected by Wigner RCP.