Nonlinear filtering in digtal image processing
In this thesis an application named ImageLab have been developed. Image-
Lab is an image filtering application focusing on the use of nonlinear filtering
processes. During development of the software it became clear that ImageLab
is not a full image processing suite, such as Photoshop, but rather a complement
to existing software solutions. This is true for now, but further development
of the software may lead it to become an alternative full featured image processing
Theory behind the use of ImageLabs image processing techniques was introduced
in section 2, the same techniques was also introduced as pseudocode.
Also the different filters and their functionality were explained. In section
4.2 the same pseudo-code were turned into usable C#-code. Following this
recipe it should be a straightforward process to duplicate the work done in this
thesis, both for novel applications or extending the current application further.
In section 5 it was shown how the different filters performed with respect to
an objective metric. While not being able to generally conclude a given filters
proficiency at removing noise, we can conclude that for a given set of initial conditions (input image, the type/strength of noise and parameters) some filters did
perform better than others. With 2-sided variable height impulsive noise and
the Lena reference image a median-filter and a Distance Weighted Median-filter
performed the best (according to the objective Mean Square Error). All the filters
did improve the objective metric, the worst (Mean 11-by-11) was able to
remove more than 40%4 of the noise. The best filter (Median 1-by-1) was able
to remove more than 90%5 of the noise.
Place, publisher, year, edition, pages
Institutt for elektronikk og telekommunikasjon , 2011. , 49 p.
ntnudaim:4673, MTEL elektronikk, Signalbehandling og kommunikasjon
IdentifiersURN: urn:nbn:no:ntnu:diva-15009Local ID: ntnudaim:4673OAI: oai:DiVA.org:ntnu-15009DiVA: diva2:473496
Ramstad, Tor Audun, Professor