Low-resolution Panchromatic Image Research Articles

Pansharpening based on total least squares regression and ratio enhancement

ABSTRACT The key of ratio enhancement (RE) is to synthesize a low-resolution panchromatic image. The grey-level distortion of the synthesized low-resolution image can distort a fused image. To tackle this problem, we propose a pansharpening method based on total least-squares regression and ratio enhancement (TLSR-RE). This method can correct the grey-level distortion in the low-resolution panchromatic synthetic image for different ground objects. The normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) are adopted to classify the pixels into different classes. Then, we employe total least squares (TLS) regression to obtain the error-free panchromatic and multispectral image of each class. The low-resolution panchromatic image is synthesized by weighted summation of the error-free multispectral image. In addition, the grey-level distorted pixels are extracted for further correction. Finally, the multispectral image is sharpened by the ratio enhancement method. The experimental results reveal that the proposed algorithm achieves high-fidelity in spatial details and spectrum.

Pansharpening by interspectral similarity and edge information using improved deep residual network

Remote sensing image pansharpening involves the fusing of multispectral (MS) images with a panchromatic (PAN) image to produce an image with high-spatial as well as high-spectral resolution. We propose an improved pansharpening algorithm based on deep learning. A four-layer residual network is used as a reconstructed model to enable the accurate estimation of high-frequency details. We consider two priors to take advantage of MS information. The first prior indicates interspectral similarity, wherein the relationship between high- and low-resolution PAN images is used in the estimation of high-resolution MS images. The second prior provides the location of edges and textures according to the gradient of the PAN image. Consistency in the spectral characteristic is used as the basis in creating a pretrained model with the aim of accelerating convergence. Multiple evaluation metrics were applied to simulated and real images in order to compare the efficacy of the proposed method with that of state-of-the-art image fusion methods.

Revealing Implicit Assumptions of the Component Substitution Pansharpening Methods

The component substitution (CS) pansharpening methods have been developed for almost three decades and have become better understood recently by generalizing them into one framework. However, few studies focus on the statistical assumptions implicit in the CS methods. This paper reveals their implicit statistical assumptions from a Bayesian data fusion framework and suggests best practices for histogram matching of the panchromatic image to the intensity image, a weighted summation of the multispectral images, to better satisfy these assumptions. The purpose of histogram matching was found to make the difference between the high-resolution panchromatic and intensity images as small as possible, as one implicit assumption claims their negligible difference. The statistical relationship between the high-resolution panchromatic and intensity images and the relationship between their corresponding low-resolution images are the same, as long as the low resolution panchromatic image is derived by considering the modulation transfer functions of the multispectral sensors. Hence, the histogram-matching equation should be derived from the low-resolution panchromatic and intensity images, but not derived from the high-resolution panchromatic and expanded low-resolution intensity images. Experiments using three example CS methods, each using the two different histogram-matching equations, was conducted on the four-band QuickBird and eight-band WorldView-2 top-of-atmosphere reflectance data. The results verified the best practices and showed that the histogram-matching equation derived from the high-resolution panchromatic and expanded low-resolution intensity images provides more-blurred histogram-matched panchromatic image and, hence less-sharpened pansharpened images than that derived from the low-resolution image pair. The usefulness of the assumptions revealed in this study for method developers is discussed. For example, the CS methods can be improved by satisfying the assumptions better, e.g., classifying the images into homogenous areas before pansharpening, and by changing the assumptions to be more general to address their deficiencies.

A Robust Image Fusion Method Based on Local Spectral and Spatial Correlation

To solve the potential color distortion problem of synthetic-variable-ratio method, an improved fusion method based on local spectral and spatial correlation (SSC) is presented. The proposed method, which uses SSC characteristics and local optimization strategy to simulate a low-resolution panchromatic image, can effectively reduce the spectral distortion of the fused image. QuickBird and other satellite images are used to assess the quality of the method. Visual and quantitative analysis demonstrates that the proposed approach can significantly improve the fusion quality.

Introducing contrast and luminance normalisation to improve the quality of subtractive resolution merge technique

Subtractive resolution merge (SRM) is a contemporary image fusion method that produces highly preserved spatial and spectral resolution. This method is limited to dual sensor platforms with specific band ratios between the high-resolution panchromatic image (HRPI) and the low-resolution multispectral image (LRMI). An additional problem with SRM is that some bands are over- or under-represented due to the normalisation function applied. This article provides two modifications that resolve these limitations. SRM builds a synthetic low-resolution panchromatic image (LRPISYN) from the weighted sum of the LRMI bands. This image is modified by using a spatially resampled HRPI instead. The second modification is the use of a contrast and luminance index for the normalising function. These two modifications are tested on QuickBird images (multispectral and panchromatic), as well as fusing SPOT-5 (Satellite Pour l'Observation de la Terre) multispectral image and an aerial photograph. The results show improved quantitative metrics and unsupervised classification compared with the standard SRM technique and other contemporary image fusion methods. Both of these modifications are grouped into a patent pending technique that is called contrast and luminance normalised fusion.

Pansharpening Through Proportional Detail Injection Based on Generalized Relative Spectral Response

In this letter, we propose a novel pansharpening method which enhances the spatial resolution of low-spatial-resolution multispectral (MS) images through proportional detail injection (PDI) based on the generalized relative spectral response (GRSR). In the algorithm, the spatial details are extracted as the difference between the panchromatic (PAN) image and the resampled low-resolution PAN image. Then, the spatial details are further modulated and injected into each band of the resampled MS image according to the proportion of the intensity of the low-frequency component of each band to the offset-removed synthetic intensity component obtained according to the GRSR. Both quantitative and qualitative evaluations demonstrate the superiority of the proposed method in terms of spatial enhancement, spectral fidelity, and computational complexity.

Simulation of Low-Resolution Panchromatic Images by Multivariate Linear Regression for Pan-Sharpening IKONOS Imageries

The extraction of spatial details is crucial for fusion quality. An efficient way is to exploit the difference between high-resolution panchromatic (Pan) images and low-resolution Pan (LRP), which is to be simulated by weighted average value from low-resolution multispectral images. To obtain the weighting coefficients with multivariate linear regression, three issues were discussed, and corresponding solutions were proposed in this letter. The proposed method consists of separating high-frequency pixels from low-frequency pixels using support vector machine and selecting observations that are evenly distributed by a bucketing technique and forcing coefficients to be sound physically by constrained least squares. Validation experiments are undertaken using three IKONOS data sets, and fusion results are compared against four popular methods. The results show that the proposed method can simulate LRP soundly and therefore achieve a better fusion quality.

M-band wavelet transform for the generation of high-resolution multispectral images

This paper introduces an image fusion method, multiresolution analysis-based intensity modulation (MRAIM), which can be used to fuse a high-resolution panchromatic image (HRPI) with low-resolution multispectral images (LRMIs) to yield high-resolution multispectral images (HRMIs). It is designed to solve the fusion case in which the resolution ratio between the LRMIs and the HRMI is an arbitrary integer. It is based on the physical principle of remote sensing image formation, the M-band wavelet transform, and the à trous algorithm. A general image fusion method is derived from the physical principle of image formation, which in turn is based on the simplified radiation transfer model. Using the general image fusion method, it is shown that the pixel values of HRMIs are determined by the pixel values of a low-resolution panchromatic image (LRPI), the approximation of the HRPI at the resolution level of the LRMIs. The M-band wavelet transform-based à trous algorithm is used to compute the LRPI from the corresponding HRPI. The key problem in implementing this approach is constructing the M-band à trous low-pass filter. The underlying mathematical concepts of the M-band wavelet transform, construction of a K-regularity M-band scaling filter, and the à trous algorithm are briefly explained. The performance of the MRAIM method is compared with a state-of-the-art wavelet transform image fusion method, AWRGB, in an experiment using IKONOS panchromatic and multispectral images. It is shown that the MRAIM method can yield a higher quality synthesis of the HRMIs and can be used to generate IKONOS HRMIs.

A comparative analysis of image fusion methods

There are many image fusion methods that can be used to produce high-resolution multispectral images from a high-resolution panchromatic image and low-resolution multispectral images. Starting from the physical principle of image formation, this paper presents a comprehensive framework, the general image fusion (GIF) method, which makes it possible to categorize, compare, and evaluate the existing image fusion methods. Using the GIF method, it is shown that the pixel values of the high-resolution multispectral images are determined by the corresponding pixel values of the low-resolution panchromatic image, the approximation of the high-resolution panchromatic image at the low-resolution level. Many of the existing image fusion methods, including, but not limited to, intensity-hue-saturation, Brovey transform, principal component analysis, high-pass filtering, high-pass modulation, the a/spl grave/ trous algorithm-based wavelet transform, and multiresolution analysis-based intensity modulation (MRAIM), are evaluated and found to be particular cases of the GIF method. The performance of each image fusion method is theoretically analyzed based on how the corresponding low-resolution panchromatic image is computed and how the modulation coefficients are set. An experiment based on IKONOS images shows that there is consistency between the theoretical analysis and the experimental results and that the MRAIM method synthesizes the images closest to those the corresponding multisensors would observe at the high-resolution level.