Review and Performance Comparison of Pansharpening Algorithms for RASAT Images

Abstract

This study presents the most extensive performance comparison of pansharpening methodologies by considering 17 pansharpening algorithms that are applied to the satellite images obtained from RASAT, which is the first earth observation satellite designed and manufactured in Turkey. Standard and state-of-the-art pansharpening approaches from various categories, such as component substitution (CS), modulation based (MB), multiresolution analysis (MRA), and hybrid and variational methods, are included in order to gain a better insight and perform a thorough analysis of the performance of various pansharpening methods. The experimental validation procedure was designed according to Wald's protocol, and the performance evaluations were conducted both qualitatively and quantitatively on the basis of seven quantitative evaluation criteria.

Hybrid Attention Based Residual Network for Pansharpening

“…Existing pansharpening methods fall into two main categories: traditional fusion algorithms and deep learning (DL) based methods, among which the former is divided into component substitution algorithms (CS), multi-resolution analysis algorithms (MRA), and sparse representation (SR) based algorithms [3][4][5][6].…”

“…However, the fusion results of GS-SRCNN still suffer from spectral distortion and lack of spatial details. Creatively, Masi et al [10] proposed the PNN model based on convolutional neural network (CNN) [39] for the first time, which is composed of three convolution layers using kernel size (9,5,5). To improve the fusion accuracy, this same lab introduced nonlinear radiation index into PNN [40], yet the fused images still contained spectral distorted pixels and unclear edges, which implies the limitation of single-branch networks.…”

“…However, due to the restrictions of technical conditions and hardware limitations [2], optical remote sensing satellites can only provide high-resolution PAN images and low-resolution MS images. PAN image has only one spectral channel, which means it cannot express RGB colors, and on the contrary, MS image carries high expression ability of color [3][4][5][6]. Therefore, the fusion of high spatial resolution of PAN and high spectral resolution of MS, called pansharpening, is proposed and proven to be an effective method.…”

Variational Bayesian Pansharpening with Super-Gaussian Sparse Image Priors

“…Over the past 20 years, numerous methods have been presented and, in an attempt to bring some order to the diversity of approaches, different reviews, comparisons and classifications have been proposed in the literature (see, for instance, [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 ]) each one with different criteria and, therefore, with a different categorization. Nevertheless, in the last years, there seems to be a consensus in three main categories, namely Component Substitution (CS), Multi-Resolution Analysis (MRA) and Variational Optimization (VO) [ 15 , 16 , 17 ]. Additionally, the increasing number of Deep Learning (DL)-based pansharpening methods proposed in recent years can be regarded as a new category.…”

Fusion of High-Resolution SAR and Optical Imageries Based on a Wavelet Transform and IHS Integrated Algorithm

“…The Pyramid-based image fusion methods, including Laplacian pyramid transform, were all developed from Gaussian pyramid transform which has been modified and widely used in a number of image processing applications including the image fusion [10] [11]. Particularly, The MRA fusion approaches usually to provide superior performance in terms of reducing the spectral distortion compared to CS or spatial domain based methods [12] [13]. On the other hand PCA transformation can acquire higher spatial resolution but provides more serious distortion of spectral characteristics.…”