Design of the Image Denoising Tool

Travail de diplome en informatique
Computer Vision & Multimedia Laboratory
Candidat : ???
Responsables : Sviatoslav Vololshynovskiy
Description

 Denoising of images should be carried out when the quality of observed visual data is not sufficient for next image processing stages. Known techniques
 are either very complex, parameterized, use nontractable models and computationally extensive or.  Therefore it is  necessary to develop methods that 
 are able to overcome these drawbacks.
 

 In the scope of diploma project, a successful candidate will help us to implement denoising technique based on the nonstationary stochastic image 
 model based on existing Matlab code. It will be necessary to develop and to integrate denoising software into the existing interface which allows to easily  
 link dll's and supports more then 20 graphical formats.
 
Outline
  • programming in C/C++ or Java
  • basic knowledge of image processing techniques
  • evaluating the effectiveness of the developed software
References

| Research | Members | Publications | Software (LaboImage) | Demonstrations | Teaching | 

| Diploma and licence subjects | Jobs and Postgraduate Opportunities | Search | Home | 

Computer Vision Group 
University of Geneva 
24 rue du Général-Dufour 
1211 Geneva 4 
Switzerland 

Group director : 
Thierry Pun

Designed by : 
Lori Stefano Petrucci and David Squire

Generated from XML source and an XSL stylesheet, using XSL:P


By Stéphane Marchand-Maillet
_______________________ 
Copyright © 1999 CUI