Abstract

The problem of the synthesis of an adaptive nonlinear algorithm for filtering a sequence of images utilized for achieving superresolution under the conditions of unknown parameters accompanying the process of observation is considered. Using the method of partitioning, algorithms for adaptive filtering in a block form are developed and studied. In this approach, adaptation is performed with respect to unknown values of the parameters of interframe displacements (parameters of an affine transformation). The quality of reconstructed images is improved compared with the reconstruction quality using known algorithms based on the use of fixed estimates of the displacements over the images of the processed sequence with low resolution.

© 2017 Optical Society of America

Superresolution image reconstruction from a sequence of aliased imagery

S. Susan Young and Ronald G. Driggers
Appl. Opt. 45(21) 5073-5085 (2006)

Data-driven multichannel superresolution with application to video sequences

H. Shekarforoush and R. Chellappa
J. Opt. Soc. Am. A 16(3) 481-492 (1999)

Spectral-overlap approach to multiframe superresolution image reconstruction

Edward Cohen, Richard H. Picard, and Peter N. Crabtree
Appl. Opt. 55(15) 3978-3992 (2016)

Image preprocessing for improving computational efficiency in implementation of restoration and superresolution algorithms

Malur K. Sundareshan, Supratik Bhattacharjee, Radhika Inampudi, and Ho-Yuen Pang
Appl. Opt. 41(35) 7464-7474 (2002)

Spatial-frequency-based metric for image superresolution

Matthew Woods and Aggelos K. Katsaggelos
J. Opt. Soc. Am. A 32(11) 2002-2020 (2015)