Image Deblurring Via an Adaptive Dictionary Learning Strategy
Abstract: Recently, sparse
representation has been applied to image deblurring. The dictionary is the fundamental
part of it and the proper selection of dictionary is very important to achieve
superperformance. The global learned dictionary might achieve inferior
performances since it could not mine the specific informati n such as the
texture and edge which is contained in the blurred image. However, it is a computational
burden to train a new dictionary for image deblurring which requires the whole
image (or most parts)as input; training the dictionary on only a few patches
would result in over-fitting. To addressthe problem, we instead propose an
online adaption strategy to transfer the global learned dictionary to a specific
image. In our deblurring algorithm, the sparse coefficients, latent image, blur
kernel and the dictionary are updated alternatively. And in every step, the
global learned dictionary is updated in an online form via sampling only few
training patches from the target noisy image. Since our adaptive dictionaryexploits
the specific information, our deblurring algorithm shows superior performance
over otherstate-ofthe-art algorithms.
Author: Lei Li, Ruiting Zhang
Journal Code: jptkomputergg140120
