A performance evaluation of image interpolation and superresolution algorithms

Author

Ye, Xin ; Lu, Xiqun

Author_Institution

Coll. of Comput. Sci., Zhejiang Univ., Hangzhou, China

fYear

2011

fDate

26-28 July 2011

Firstpage

4776

Lastpage

4779

Abstract

Image interpolation and superresolution can be considered as a problem of estimating a high-resolution image from a single or multiple registered low-resolution images. In this paper we classify image interpolation and super-resolution estimators into linear and nonlinear, fixed and signal adaptive model, local and non-local, learning and non-learning-based, back-projection and non-back-projection. We compare the performance of different estimators for various gray and color images by using both objective measurements, such as MSSIM and PSNR, and subjective observation.

Keywords

backpropagation; image colour analysis; image registration; image resolution; interpolation; learning (artificial intelligence); performance evaluation; backprojection; color images; fixed model; gray images; image interpolation; learning-based model; linear model; multiple registered low-resolution images; nonbackprojection; nonlearning-based model; nonlinear model; objective measurements; performance evaluation; signal adaptive model; single registered low-resolution images; subjective observation; superresolution algorithms; Classification algorithms; Image edge detection; Image reconstruction; Interpolation; Signal resolution; Spatial resolution; backprojection; image interpolation; local statistical; non-olcal; superresolution;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia Technology (ICMT), 2011 International Conference on

Conference_Location

Hangzhou

Print_ISBN

978-1-61284-771-9

Type

conf

DOI

10.1109/ICMT.2011.6003023

Filename

6003023