Least-squares Optimal Interpolation for Fast Image Super-resolution

Image super-resolution is generally regarded as consisting of three steps – image registration, fusion, and deblurring. This paper presents a novel technique for resampling a non-uniformly sampled image onto a uniform grid that can be used for fusion of translated input images. The proposed method can be very fast, as it can be implemented as a finite impulse response filter of low order (10th order results in good performance). The technique is based on optimising the resampling filter coefficients using a simple image model in a least squares fashion. The method is tested experimentally on a range of images and shown to have similar results to that of a least-squares optimal filter. Further experimental comparisons are made against a number of methods commonly used in image super-resolution that show that the proposed method is superior to these.

[1]  Michael Unser,et al.  Image interpolation and resampling , 2000 .

[2]  Moon Gi Kang,et al.  Super-resolution image reconstruction: a technical overview , 2003, IEEE Signal Process. Mag..

[3]  Klamer Schutte,et al.  Robust Fusion of Irregularly Sampled Data Using Adaptive Normalized Convolution , 2006, EURASIP J. Adv. Signal Process..

[4]  D. Shepard.  A two-dimensional interpolation function for irregularly-spaced data , 1968, ACM National Conference.

[5]  Michael Elad,et al.  Restoration of a single superresolution image from several blurred, noisy, and undersampled measured images , 1997, IEEE Trans. Image Process..

[6]  N. K. Bose,et al.  High resolution image formation from low resolution frames using Delaunay triangulation , 2002, IEEE Trans. Image Process..

[7]  M. Unser.  Sampling-50 years after Shannon , 2000, Proceedings of the IEEE.

[8]  Russell C. Hardie,et al.  Joint MAP registration and high-resolution image estimation using a sequence of undersampled images , 1997, IEEE Trans. Image Process..

[9]  Carlos Vázquez,et al.  Reconstruction of nonuniformly sampled images in spline spaces , 2005, IEEE Transactions on Image Processing.

[10]  Stephen R. Marsland,et al.  Interpolation Models for Image Super-resolution , 2008, 4th IEEE International Symposium on Electronic Design, Test and Applications (delta 2008).

[11]  Andrew Gilman.  Least-squares optimal interpolation for direct image super-resolution : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Engineering at Massey University, Palmerston North, New Zealand , 2009 .

[12]  Donald G. Bailey,et al.  From , 2020, The Solace Is Not the Lullaby.