Super-resolution algorithm based on sparse representation and wavelet preprocessing for remote sensing imagery

Ruizhi Ren; Lingjia Gu; Haoyang Fu; Chenglin Sun

doi:10.1117/1.JRS.11.026014

10 May 2017 Super-resolution algorithm based on sparse representation and wavelet preprocessing for remote sensing imagery

Ruizhi Ren, Lingjia Gu, Haoyang Fu, Chenglin Sun

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 11, Issue 2, 026014 (May 2017). https://doi.org/10.1117/1.JRS.11.026014

Abstract

An effective super-resolution (SR) algorithm is proposed for actual spectral remote sensing images based on sparse representation and wavelet preprocessing. The proposed SR algorithm mainly consists of dictionary training and image reconstruction. Wavelet preprocessing is used to establish four subbands, i.e., low frequency, horizontal, vertical, and diagonal high frequency, for an input image. As compared to the traditional approaches involving the direct training of image patches, the proposed approach focuses on the training of features derived from these four subbands. The proposed algorithm is verified using different spectral remote sensing images, e.g., moderate-resolution imaging spectroradiometer (MODIS) images with different bands, and the latest Chinese Jilin-1 satellite images with high spatial resolution. According to the visual experimental results obtained from the MODIS remote sensing data, the SR images using the proposed SR algorithm are superior to those using a conventional bicubic interpolation algorithm or traditional SR algorithms without preprocessing. Fusion algorithms, e.g., standard intensity-hue-saturation, principal component analysis, wavelet transform, and the proposed SR algorithms are utilized to merge the multispectral and panchromatic images acquired by the Jilin-1 satellite. The effectiveness of the proposed SR algorithm is assessed by parameters such as peak signal-to-noise ratio, structural similarity index, correlation coefficient, root-mean-square error, relative dimensionless global error in synthesis, relative average spectral error, spectral angle mapper, and the quality index Q4, and its performance is better than that of the standard image fusion algorithms.

Citation Download Citation

Ruizhi Ren, Lingjia Gu, Haoyang Fu, and Chenglin Sun "Super-resolution algorithm based on sparse representation and wavelet preprocessing for remote sensing imagery," Journal of Applied Remote Sensing 11(2), 026014 (10 May 2017). https://doi.org/10.1117/1.JRS.11.026014

Received: 19 November 2016; Accepted: 19 April 2017; Published: 10 May 2017

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