Brandt, Rafael (2018) Evaluation of Deconvolution Methods. Master's Internship Report, Computing Science.
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (94kB) |
|
|
Text
internshipbrandt.pdf Download (3MB) | Preview |
Abstract
Where blind deconvolution is the recovery of a true image from a (noisy) convolved image, non-blind deconvolution is the recovery of a true image from the point spread function it is convolved with and a (noisy) convolved image. Four blind and four non-blind deconvolution methods were assessed and compared. Experimental evaluation was performed with different noise types (i.e. shotnoise, Gaussian noise and impulse noise), noise intensities, point spread function types (i.e. out-of-focus blur, Gaussian blur, linear motion blur and non-linear motion blur), point spread function sizes, and true images. The results of the said empirical evaluation were presented in this document. Correlations were found between variables (such as method, noise type, noise intensity, etc.) and reconstruction quality as well as runtime and memory usage. The in this document presented results indicated that the use case dictates which deconvolution method is appropriate.
| Item Type: | Thesis (Master's Internship Report) |
|---|---|
| Supervisor name: | Wilkinson, M.H.F. and Biehl, M. |
| Degree programme: | Computing Science |
| Thesis type: | Master's Internship Report |
| Language: | English |
| Date Deposited: | 18 Jul 2018 |
| Last Modified: | 07 Aug 2018 10:05 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/17930 |
Actions (login required)
 |
View Item |