Advanced

Reconstruction of NDVI Data with Normal Variance-Mean Mixture Noise using Stochastic Gradient Methods

Kraft, David (2017) FMS820 20171
Mathematical Statistics
Abstract
Satellite data is useful for making inference on large areas, but there are plenty of problems associated with the technology. The observation errors need to be carefully modelled in order to make inference from the data. In addition, the large data-sets involved makes straight-forward computations infeasible.
This thesis explores a non-Gaussian noise structure, and implements a modified EM algorithm where the M-step is replaced for taking a step in a stochastic gradient descent method. The estimated parameters are used for reconstruction both of simulated data and of a data-set retrieved from the PAL database.
The algorithm performs well on a simulated data equipped with the non-Gaussian noise structure; as for the PAL data, the... (More)
Satellite data is useful for making inference on large areas, but there are plenty of problems associated with the technology. The observation errors need to be carefully modelled in order to make inference from the data. In addition, the large data-sets involved makes straight-forward computations infeasible.
This thesis explores a non-Gaussian noise structure, and implements a modified EM algorithm where the M-step is replaced for taking a step in a stochastic gradient descent method. The estimated parameters are used for reconstruction both of simulated data and of a data-set retrieved from the PAL database.
The algorithm performs well on a simulated data equipped with the non-Gaussian noise structure; as for the PAL data, the algorithm performed quite well, but convergence proved to be somewhat unstable. (Less)
Please use this url to cite or link to this publication:
author
Kraft, David
supervisor
organization
course
FMS820 20171
year
type
H2 - Master's Degree (Two Years)
subject
language
English
id
8902621
date added to LUP
2017-02-07 08:19:48
date last changed
2017-02-07 08:19:48
@misc{8902621,
  abstract     = {Satellite data is useful for making inference on large areas, but there are plenty of problems associated with the technology. The observation errors need to be carefully modelled in order to make inference from the data. In addition, the large data-sets involved makes straight-forward computations infeasible.
This thesis explores a non-Gaussian noise structure, and implements a modified EM algorithm where the M-step is replaced for taking a step in a stochastic gradient descent method. The estimated parameters are used for reconstruction both of simulated data and of a data-set retrieved from the PAL database.
The algorithm performs well on a simulated data equipped with the non-Gaussian noise structure; as for the PAL data, the algorithm performed quite well, but convergence proved to be somewhat unstable.},
  author       = {Kraft, David},
  language     = {eng},
  note         = {Student Paper},
  title        = {Reconstruction of NDVI Data with Normal Variance-Mean Mixture Noise using Stochastic Gradient Methods},
  year         = {2017},
}