A New Pan-Sharpening Method With Deep Neural Networks

By Liang Xiao, Zhihui Wei, Hongyi Liu, Songze Tang. IEEE Geoscience and Remote Sensing Letters, 2015

303 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/113382d800e2d9cc269a29d96c9b288fc50e2a7e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Boosting the Accuracy of Multispectral Image Pansharpening by Learning a Deep Residual Network (2017, result): “Similar to the successful applications for the single-image super-resolution problem [9], [10], deep learning models have also been introduced to the field of image fusion [11], [12].”
PanNet: A Deep Network Architecture for Pan-Sharpening (2017, methodology): “as that between the corresponding HR/LR panochromatic image patches, and uses this assumption to learn a mapping through a neural network [16].”
Using convolutional features and a sparse autoencoder for land-use scene classification (2016, unknown): “Huang et al. (2015) proposed a DNN-based method for pan-sharpening.”
Target-Adaptive CNN-Based Pansharpening (2017, background): “Very recently, following the recent technological and theoretical advances in computer vision and related fields, also deep learning methods have been applied to remote sensing problems [37], [38], [39], and several papers have been proposed to address pansharpening [40], [41], [42], [43], [44], [45”
Pansharpening by Convolutional Neural Networks (2016, background): “As of today, instead, there has been limited interest on deep learning, with some exception like in [33], where a modiﬁed sparse denoising autoencoder algorithm is proposed.”
Deep Learning Meets Hyperspectral Image Analysis: A Multidisciplinary Review (2019, methodology): “Conversely, in [82] modiﬁed sparse denoising AEs were used to train a mapping between low-resolution and high-resolution image patches for pan-sharpening.”
Deep Learning for Image Super-Resolution: A Survey (2019, background): “Among them, Huang et al. [178] present a sparse denoising autoencoder to learn LR-to-HR mappings with PANs and transfer it to HSIs.”
Deep Learning in Multimodal Remote Sensing Data Fusion: A Comprehensive Review (2022, methodology): “Huang et al. (2015) propose the first DL-based method in dealing with pansharpening problem, where a sparse denoising AE is adopted to learn the transformation in Pan domain, and then the observed MS is input into the pretrained AE to generate final output.”
Comprehensive survey of deep learning in remote sensing: theories, tools, and challenges for the community (2017, methodology): “Huang et al.253 use a modified sparse denoising autoencoder (SPDAE), denoted MSDAE, which uses the SPDAE to represent the relationship between the HR image patches as clean data to the lower spatial resolution, high spectral resolution MSI image as corrupted data.”
Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art (2016, methodology): “For example, for RS image pan sharpening, they are the HR and low-resolution (LR) image patches from the panchromatic (PAN) images [28]; for pixel-based classification, they are the spectral–spatial features and their feature representations (unsupervised version) or label information (supervised ve”
Detail Injection-Based Deep Convolutional Neural Networks for Pansharpening (2021, background): “With the tremendous improvements of hardware, convolutional neural networks (CNNs) have recently become a powerful tool to deal with pansharpening and its related applications (see [1], [2], [40]–[54]).”
Deep learning in remote sensing: a review (2017, methodology): “ures, which have been trained on massive datasets with very big models, is often a relevant solution, since one makes use of discriminative strong features and only injects task-speciﬁc knowledge. In [125], authors learn an upsampling of the panchromatic band via a stack of autoencoders: the model i”
Multiview Deep Learning for Land-Use Classification (2015, methodology): “[12] using stacked modified sparse denoising autoencoders for pretraining the hidden layers of the DNN to avoid the “diffusion of gradients” caused by random neuron initialization.”
A Multiscale and Multidepth Convolutional Neural Network for Remote Sensing Imagery Pan-Sharpening (2017, background): “Examples are the sparse deep neural network [32] and the pan-sharpening neural network (PNN) [33],”
Multispectral and Hyperspectral Image Fusion Using a 3-D-Convolutional Neural Network (2017, methodology): “DL-based methods have previously been used to solve the pansharpening problem [13], [14].”