
ndSuperRes
This is a page dedicated to ndSuperRes, the super resolution demo for Nanodesktop PSPE/PSP.
Super-resolution (SR) are techniques that in some way enhance the
resolution of an imaging system.
There are different views as to what is considered an SR technique: some consider only techniques that break the diffraction-limit of systems, while others also consider techniques that merely break the limit of the digital imaging sensor as SR.
Filippo Battaglia at Visilab Research Center of the University of Messina has
demonstrated the possibility of executing a SR algorithm on a simple embedded
device as the Sony Playstation Portable. This result shall be possible through
the use of Nanodesktop SDK.
This program uses the super resolution algorithm developed by Michael Smith at
Carlenton University (http://www.engsoc.org/~michael/).
You can find more informations about the internal implementation of Smith's algorithm in the current Nanodesktop distribution, in the /ndSuperRes folder, or in Mr. Smith's website.
Download SuperRes program vers. 1.1
A screenshot of the program:

We can see shortly how the program works. ndSuperRes generates a single high resolution image starting from n low-resolution images (partial frames) of the same object. See this picture:

As you can see, the program has merged the informations coming from 16 frames of 135x111 pixels, and it has generated a single high resolution image (in the right side of the screen).
The output image has a doubled resolution compared with the partial input images (at the left side of the screen you can see one of these 16 frames).
But the output image appears more resolved than the 16 input images, despite it has a larger resolution (270x222 pixels).
Università degli
Studi di Messina
Visilab
Computer Vision Laboratory
