Enhanced anchored neighborhood regression for image super resolution based on maximized utilization on dictionary atoms and training sample patches

Qinfen Yao

doi:10.1117/12.2587039

12 March 2021 Enhanced anchored neighborhood regression for image super resolution based on maximized utilization on dictionary atoms and training sample patches

Qinfen Yao

Author Affiliations +

Proceedings Volume 11763, Seventh Symposium on Novel Photoelectronic Detection Technology and Applications; 117634P (2021) https://doi.org/10.1117/12.2587039
Event: Seventh Symposium on Novel Photoelectronic Detection Technology and Application 2020, 2020, Kunming, China

Abstract

The two recent effective image super resolution algorithms, Anchored Neighborhood Regression (ANR) and its improved variation, Adjusted ANR (A+), have provided state-of-the-art performance and fast execution speed. In this paper, we further explored the improvable points of A+ and proposed the enhanced ANR (EA) algorithm. First, for a better approximation of input image patches, EA learns the dictionary with least sparsity limit to ensure more patches adhere to these atoms. Second, EA uses the correlation rather than the Euclidean distance to measure the similarity between training sample patches and dictionary atoms when searching neighbors. Third, sample patches are not only clustered to compute the regressors, but also seen as potential atoms in our EA. These potential atoms work together with dictionary atoms, help find the atom most similar with the input image patch as far as possible. We demonstrate the results on commonly used datasets, showing both better visual performance and higher index values compared to state-of-the-art methods.

Citation Download Citation

Qinfen Yao "Enhanced anchored neighborhood regression for image super resolution based on maximized utilization on dictionary atoms and training sample patches", Proc. SPIE 11763, Seventh Symposium on Novel Photoelectronic Detection Technology and Applications, 117634P (12 March 2021); https://doi.org/10.1117/12.2587039

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available