M.Sc Thesis


M.Sc StudentLoeub Tamar
SubjectHierarchical Scattering Tomography with a
Monotonicity Prior
DepartmentDepartment of Electrical and Computer Engineering
Supervisor PROF. Yoav Schechner
Full Thesis textFull thesis text - English Version


Abstract

We enable large scale analysis of three-dimensional heterogeneous scattering fields, through stochastic scattering tomography which relies on radiative transfer.
This is done in a coarse-to-fine (hierarchical) approach. The approach contrasts with state-of-the-art recovery, which is based on memory-limited deterministic radiative transfer and thus applicable to smaller domains. 

Tomography is achieved via stochastic gradient descent, where the gradient is derived via Monte Carlo.

We introduce a differentiable monotonicity prior, useful to express signals of monotonic tendency, such as the extinction coefficient in clouds, as a function of altitude. When deriving such a tomography approach, the case study of estimating cloud structure is important for climate research and consistent with future spaceborne imaging technologies.