A Competent Convolutional Sparse Representation Model for Pan-Sharpening of Multi-Spectral Images

Abstract

Two types of images are produced by Earth observation satellites, each having complementing spatial and
spectral characteristics. Pan-sharpening (PS) is based on remote sensing and image fusion approach that
produces a high spatial resolution multi-spectral image by merging spectral information from a low spatial
resolution multispectral (MS) image with intrinsic spatial details from a high spatial resolution panchromatic
(PAN) image. Traditional pan-sharpening methods continue to seek for a fused image that contains the
necessary spatial and spectral information. This work proposes a pan-sharpening method based on a recent
invention, convolutional sparse representation (CSR). Geometric structural characteristics are extracted from
the PAN image using a CSR-based filtering procedure. The challenge of learning filters, convolutional basis
pursuit denoising (CBPDN), is handled using a modified dictionary learning method based on the concept of
Alternating Direction Method of Multipliers (ADMM). The retrieved details are put into MS bands using
applicable weighting coefficients. Because the proposed fusion model avoids the standard patch-based
method, spatial and structural features are preserved while spectral quality is maintained. The spectral
distortion index SAM and the spatial measure ERGAS improve by 4.4 and 6.2 percent, respectively, when
compared to SR-based techniques. The computational complexity is reduced by 200 seconds when compared

to the most recent SR-based fusion technique. The proposed method's efficacy is demonstrated by reduced-
scale and full-scale experimental findings utilising the QuickBird and GeoEye-1 datasets.