Premium
P‐5.3: A super resolution reconstruction algorithm based on spatial autoregression regularization
Author(s) -
shen Huaming,
Xu Meihua,
Ran Feng,
Li Liming
Publication year - 2018
Publication title -
sid symposium digest of technical papers
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.351
H-Index - 44
eISSN - 2168-0159
pISSN - 0097-966X
DOI - 10.1002/sdtp.12789
Subject(s) - regularization (linguistics) , autoregressive model , computer science , algorithm , image resolution , artificial intelligence , image (mathematics) , sparse approximation , pattern recognition (psychology) , iterative reconstruction , computer vision , mathematics , statistics
In a lot of micro display applications, we need to enlarge and recognize a region of interest (ROI), which often get a low‐resolution image because of limited resolution source image. This paper proposed a novel image reconstruction algorithm based on spatial autoregression regularization and sparse representation. This reconstruction algorithm trained an image dictionary with the same sparse coefficient by sparse K‐SVD, and then import the autoregressive regularization item to construct the objective function which can realize local adaptive control of the image. In order to obtain further clear image, the degradation model was used which can realize the global constraints. The experimental results show that the proposed algorithm has considerable effectiveness in terms of both objective measurements and visual evaluation.