Current Projects

Spatial Resolution Enhancement Key Words: Edge-directed interpolation, linear spatially varying filters, unsharp filtering, and resolution enhancement

Project Members

• Ibrahim E. Pekkucuksen
• Osman G. Sezer
• Michael J. Santoro
• Abdul G. Rana

Project Description

Spatial resolution enhancement is one of the most fundamental problems in digital image processing. Almost every digital image processing application requires some level of spatial resolution increase. Being one of the oldest research topics in the area, resolution enhancement has been thoroughly studied by many researchers. Previously proposed spatial interpolation methods include quick and dirty methods like bilinear interpolation, methods that impose constraints on the gradients like cubic interpolation and cubic spline interpolation, edge-directed methods like the NEDI algorithm, and sub-band filtering methods like the DCT and wavelet based algorithms. Our goal is to devise a linear edge-directed (i.e., spatially-varying) filter with reduced computational complexity, improved edge-detection and edge-alignment.

Related Publications

• O.G. Sezer, Y. Altunbasak, "Adaptive Boxcar/Wavelet Transform", in Proc. of SPIE Visual Communications and Image Processing Conference, VCIP – 2009, San Jose, CA. [pdf]
• Toygar Akgun, Yucel Altunbasak and R. M. Mersereau, "Superresolution reconstruction of hyperspectral images," in Proc. IEEE Int. Conf. on Acoustics Speech and Signal Processing (ICASSP), Montreal, Canada, May 2004 [pdf]
• B. K. Gunturk, Y. Altunbasak, and R. M. Mersereau, “Multi-frame information fusion for gray-scale and spatial enhancement of images,” in Proc. IEEE Int. Conf. Image Processing, Barcelona, Spain, Sep. 2003. [pdf]
• B. Gunturk, Y. Altunbasak, and R. M. Mersereau, “Gray-scale resolution enhancement,” IEEE Workshop on Multimedia Signal Processing, pp. 155-160, Cannes, France, Oct. 2001 [pdf]