Sparsity inspired pan-sharpening technique using multi-scale learned dictionary

Gogineni, Rajesh; Chaturvedi, Ashvini

doi:10.1016/j.isprsjprs.2018.10.009

Cited by 25 publications

(10 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The hierarchical characteristics of masks were used by considering various non-linear deep learning models. Gogineni and Chaturvedi [ 36 ] used multiscale learned dictionary (MSLD) to design a pan-sharpening technique. It could obtain the underlying features of images, in which the characteristics of both learned dictionaries and multiscale were possessed.…”

Section: Literature Reviewmentioning

confidence: 99%

[Retracted] Artificial Intelligence‐Based Deep Fusion Model for Pan‐Sharpening of Remote Sensing Images

Iskanderani

Mehedi

Aljohani

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

During the past two decades, many remote sensing image fusion techniques have been designed to improve the spatial resolution of the low-spatial-resolution multispectral bands. The main objective is fuse the low-resolution multispectral (MS) image and the high-spatial-resolution panchromatic (PAN) image to obtain a fused image having high spatial and spectral information. Recently, many artificial intelligence-based deep learning models have been designed to fuse the remote sensing images. But these models do not consider the inherent image distribution difference between MS and PAN images. Therefore, the obtained fused images may suffer from gradient and color distortion problems. To overcome these problems, in this paper, an efficient artificial intelligence-based deep transfer learning model is proposed. Inception-ResNet-v2 model is improved by using a color-aware perceptual loss (CPL). The obtained fused images are further improved by using gradient channel prior as a postprocessing step. Gradient channel prior is used to preserve the color and gradient information. Extensive experiments are carried out by considering the benchmark datasets. Performance analysis shows that the proposed model can efficiently preserve color and gradient information in the fused remote sensing images than the existing models.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

[Retracted] Artificial Intelligence‐Based Deep Fusion Model for Pan‐Sharpening of Remote Sensing Images

Iskanderani

Mehedi

Aljohani

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…The MRA‐based methods use multiresolution transforms (eg, wavelet, curvelet, and so on) to obtain the spatial structure of the scene and injects it into the input MS data. Although the MRA‐based methods are known for their success in preserving the color features of the MS data, they tend to deteriorate the spatial structure fidelity 1,16 . The Discrete Wavelet Transform (DWT), 17 Generalized Laplacian Pyramid (GLP) with Modulation Transfer Function (MTF) matched filter (MTF‐GLP), 18 MTF‐GLP with High‐Pass Modulation (MTF‐GLP‐HPM) 19,20 and Indusion (IND) 21 are among the most common MRA‐based methods.…”

Section: Introductionmentioning

confidence: 99%

A Non‐Dominated Sorting Genetic Algorithm‐II‐based approach to optimize the spectral and spatial quality of component substitution‐based pansharpened images

Yılmaz

2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary Pansharpening aims to fuse a lower‐resolution multispectral (MS) image and a higher‐resolution panchromatic image, resulting in an image with the color quality of the former and spatial detail quality of the latter. Of all, the component substitution (CS)‐based pansharpening methods have drawn attentions with their ability to produce sharp images. Despite their success in sharpening the images, these methods deteriorate the color features of the input MS images due of the uncertainty in the calculation of the intensity component used by them. Previous studies showed that attempts to preserve the color features tend to cause spatial detail loss to a certain extent. This, of course, reveals the necessity of a compromise between the spectral and spatial fidelity of the pansharpened images produced by the CS‐based techniques. This study proposed using the multi‐objective Non‐Dominated Sorting Genetic Algorithm‐II metaheuristic algorithm with the CS‐based methods to optimize the intensity component to find the best compromise between the spectral and spatial fidelity of the pansharpened images. The proposed framework was applied on two commonly used pansharpening techniques, Gram‐Schmidt and Synthetic Variable Ratio. It was found that the proposed methods managed to find the best balance between the color and spatial fidelity.

show abstract

“…The core idea of sparse representation is that image can be represented as linear combination of the fewest atoms in an over-complete dictionary. Some pansharpening methods based on sparse representation were proposed in [32][33][34][35][36]. Ayas et al took texture information into account in the fusion process, which protects spectra and details better [35].…”

Section: Introductionmentioning

confidence: 99%

“…Ayas et al took texture information into account in the fusion process, which protects spectra and details better [35]. Gogineni et al proposed a multi-scale learned dictionary for high frequency component [36]. Although the pansharpening methods based on sparse representation achieve good performance, they are usually time consuming.…”

Section: Introductionmentioning

confidence: 99%

Two-Stage Pansharpening Based on Multi-Level Detail Injection Network

2020

IEEE Access

View full text Add to dashboard Cite

Pansharpening is an effective technology to obtain high resolution multispectral (HRMS) images by fusing low resolution multispectral (LRMS) images and high resolution panchromatic (PAN) images. With the rapid development of deep learning, some pansharpening methods based on deep learning have been proposed. Although fused images are greatly improved, there are still some areas for improvement. For example, the spectral preservation is not good enough and the details of fused images are not rich enough. To address the above problems, a two-stage pansharpening method based on convolutional neural network (CNN) is proposed. In the first stage, image super-resolution technology with residual block is used to enhance LRMS. In order to preserve spectra, inspired by the SAM (spectral angle mapper) index, a new spectral loss function is proposed. The second stage is the fusion stage. Detail injection block is proposed by combining detail injection and CNN in this stage. Experiments on WorldView2 and GeoEye1 images demonstrate that our fused images present more spatial details and better spectra by comparing with existing methods.

show abstract

Sparsity inspired pan-sharpening technique using multi-scale learned dictionary

Cited by 25 publications

References 45 publications

[Retracted] Artificial Intelligence‐Based Deep Fusion Model for Pan‐Sharpening of Remote Sensing Images

[Retracted] Artificial Intelligence‐Based Deep Fusion Model for Pan‐Sharpening of Remote Sensing Images

A Non‐Dominated Sorting Genetic Algorithm‐II‐based approach to optimize the spectral and spatial quality of component substitution‐based pansharpened images

Two-Stage Pansharpening Based on Multi-Level Detail Injection Network

Contact Info

Product

Resources

About