Single-image super-resolution using sparse regression and natural image prior
- PMID: 20431136
- DOI: 10.1109/TPAMI.2010.25
Single-image super-resolution using sparse regression and natural image prior
Abstract
This paper proposes a framework for single-image super-resolution. The underlying idea is to learn a map from input low-resolution images to target high-resolution images based on example pairs of input and output images. Kernel ridge regression (KRR) is adopted for this purpose. To reduce the time complexity of training and testing for KRR, a sparse solution is found by combining the ideas of kernel matching pursuit and gradient descent. As a regularized solution, KRR leads to a better generalization than simply storing the examples as has been done in existing example-based algorithms and results in much less noisy images. However, this may introduce blurring and ringing artifacts around major edges as sharp changes are penalized severely. A prior model of a generic image class which takes into account the discontinuity property of images is adopted to resolve this problem. Comparison with existing algorithms shows the effectiveness of the proposed method.
Similar articles
-
Image deblurring and super-resolution by adaptive sparse domain selection and adaptive regularization.IEEE Trans Image Process. 2011 Jul;20(7):1838-57. doi: 10.1109/TIP.2011.2108306. Epub 2011 Jan 28. IEEE Trans Image Process. 2011. PMID: 21278019
-
Deblurring using regularized locally adaptive kernel regression.IEEE Trans Image Process. 2008 Apr;17(4):550-63. doi: 10.1109/TIP.2007.918028. IEEE Trans Image Process. 2008. PMID: 18390363
-
Iterative kernel principal component analysis for image modeling.IEEE Trans Pattern Anal Mach Intell. 2005 Sep;27(9):1351-66. doi: 10.1109/TPAMI.2005.181. IEEE Trans Pattern Anal Mach Intell. 2005. PMID: 16173181
-
Single image super-resolution with non-local means and steering kernel regression.IEEE Trans Image Process. 2012 Nov;21(11):4544-56. doi: 10.1109/TIP.2012.2208977. Epub 2012 Jul 16. IEEE Trans Image Process. 2012. PMID: 22829403
-
Spatially adaptive block-based super-resolution.IEEE Trans Image Process. 2012 Mar;21(3):1031-45. doi: 10.1109/TIP.2011.2166971. Epub 2011 Sep 1. IEEE Trans Image Process. 2012. PMID: 21896388
Cited by
-
FNSAM: Image super-resolution using a feedback network with self-attention mechanism.Technol Health Care. 2023;31(S1):383-395. doi: 10.3233/THC-236033. Technol Health Care. 2023. PMID: 37066938 Free PMC article.
-
Super resolution reconstruction of CT images based on multi-scale attention mechanism.Multimed Tools Appl. 2023;82(15):22651-22667. doi: 10.1007/s11042-023-14436-8. Epub 2023 Feb 7. Multimed Tools Appl. 2023. PMID: 36778717 Free PMC article.
-
Single patch super-resolution of histopathology whole slide images: a comparative study.J Med Imaging (Bellingham). 2023 Jan;10(1):017501. doi: 10.1117/1.JMI.10.1.017501. Epub 2023 Jan 31. J Med Imaging (Bellingham). 2023. PMID: 36743870 Free PMC article.
-
A Transformer-Based Model for Super-Resolution of Anime Image.Sensors (Basel). 2022 Oct 24;22(21):8126. doi: 10.3390/s22218126. Sensors (Basel). 2022. PMID: 36365830 Free PMC article.
-
Deep attention super-resolution of brain magnetic resonance images acquired under clinical protocols.Front Comput Neurosci. 2022 Aug 25;16:887633. doi: 10.3389/fncom.2022.887633. eCollection 2022. Front Comput Neurosci. 2022. PMID: 36093418 Free PMC article.
Publication types
LinkOut - more resources
Full Text Sources
Other Literature Sources
Research Materials