Image super-resolution based on hybrid multiresolution analysis

Author

Liu, Weirong ; Li, Shutao

Author_Institution

Coll. of Electr. & Inf. Eng., Hunan Univ., Changsha, China

fYear

2011

fDate

28-28 Nov. 2011

Firstpage

307

Lastpage

310

Abstract

In this paper, a novel image super-resolution scheme based on hybrid multiresolution analysis is proposed. Firstly, low resolution image is divided into components with different morphological features by morphological component analysis (MCA). Then super-resolution is performed for each component by the hybrid multiresolution analysis based on stationary wavelet transform (SWT) and nonsubsampled contourlet transform (NSCT). Experimental results on test benchmark images and license plate images validate the method´s performance.

Keywords

feature extraction; image resolution; independent component analysis; wavelet transforms; NSCT; SWT; hybrid multiresolution analysis; image super-resolution scheme; license plate image validation; morphological component analysis; morphological features; nonsubsampled contourlet transform; stationary wavelet transform; test benchmark images; Image edge detection; Image resolution; Interpolation; Multiresolution analysis; Signal resolution; Wavelet transforms; hybrid multiresolution analysis; morphological component analys; nonsubsampled contourlet transform; stationary wavelet transform; super-resolution;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition (ACPR), 2011 First Asian Conference on

Conference_Location

Beijing

Print_ISBN

978-1-4577-0122-1

Type

conf

DOI

10.1109/ACPR.2011.6166559

Filename

6166559