A Technique to Preserve Edge Information in Single Image Super Resolution

Abstract Goal of image super resolution is to enhance the size of an image without upsetting the inherited information. The quality of an enhanced image is conserved if the information around all kind of edges is preserved. The proposed novel approach potted the information around curvature edges and hard edges (abrupt transition in intensity) using Non Sub-Sampled Contourlet Transform (NSCT) based learning process. Furthermore the smoothness of smooth edges (gradual transition in intensity) is preserved by using soft edge smoothness prior as a regularizing parameter. The validity of the proposed approach is proven through simulation on several images.

[1]  Minh N. Do,et al.  Ieee Transactions on Image Processing the Contourlet Transform: an Efficient Directional Multiresolution Image Representation , 2022 .

[2]  Andrew Zisserman,et al.  Super-resolution enhancement of text image sequences , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[3]  Mei Han,et al.  SoftCuts: A Soft Edge Smoothness Prior for Color Image Super-Resolution , 2009, IEEE Transactions on Image Processing.

[4]  Amisha J. Shah,et al.  Single Image Super-Resolution via Non Sub-sample Contourlet Transform based Learning and a Gabor Prior , 2013 .

[5]  Lei Zhang,et al.  Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization , 2010, IEEE Transactions on Image Processing.

[6]  Bhabatosh Chanda,et al.  Image Upscaling Using Multiple Dictionaries of Natural Image Patches , 2012, ACCV.

[7]  David Zhang,et al.  FSIM: A Feature Similarity Index for Image Quality Assessment , 2011, IEEE Transactions on Image Processing.

[8]  Jong Beom Ra,et al.  Improvement on learning-based super-resolution by adopting residual information and patch reliability , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[9]  Debin Zhao,et al.  Image super-resolution via dual-dictionary learning and sparse representation , 2012, 2012 IEEE International Symposium on Circuits and Systems.

[10]  Guillermo Sapiro,et al.  Robust anisotropic diffusion , 1998, IEEE Trans. Image Process..

[11]  Kwang In Kim,et al.  Single-Image Super-Resolution Using Sparse Regression and Natural Image Prior , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Hua Huang,et al.  Neighbor embedding based super-resolution algorithm through edge detection and feature selection , 2009, Pattern Recognit. Lett..

[13]  Vladimir Kolmogorov,et al.  Computing geodesics and minimal surfaces via graph cuts , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[14]  Subhasis Chaudhuri,et al.  Single-Frame Image Super-resolution through Contourlet Learning , 2006, EURASIP J. Adv. Signal Process..

[15]  Thomas S. Huang,et al.  Coupled Dictionary Training for Image Super-Resolution , 2012, IEEE Transactions on Image Processing.

[16]  Aline Roumy,et al.  Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding , 2012, BMVC.

[17]  Minh N. Do,et al.  A New Contourlet Transform with Sharp Frequency Localization , 2006, 2006 International Conference on Image Processing.

[18]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.

[19]  Michael Elad,et al.  Fast and robust multiframe super resolution , 2004, IEEE Transactions on Image Processing.

[20]  Minh N. Do,et al.  Nonsubsampled contourlet transform: construction and application in enhancement , 2005, IEEE International Conference on Image Processing 2005.

[21]  Olga Veksler,et al.  Fast Approximate Energy Minimization via Graph Cuts , 2001, IEEE Trans. Pattern Anal. Mach. Intell..