• Corpus ID: 180943

Learned Spectral Super-Resolution

  title={Learned Spectral Super-Resolution},
  author={S. Galliani and Charis Lanaras and Dimitrios Marmanis and Emmanuel Baltsavias and Konrad Schindler},
We describe a novel method for blind, single-image spectral super-resolution. [] Key Method Technically, we follow the current best practice and implement a convolutional neural network (CNN), which is trained to carry out the end-to-end mapping from an entire RGB image to the corresponding hyperspectral image of equal size. We demonstrate spectral super-resolution both for conventional RGB images and for multi-spectral satellite data, outperforming the state-of-the-art.

Figures and Tables from this paper

Accurate Spectral Super-resolution from Single RGB Image Using Multi-scale CNN

A multi-scale deep convolutional neural network is presented to explicitly map the input RGB image into a hyperspectral image through symmetrically downsampling and upsampling the intermediate feature maps in a cascading paradigm, ultimately improving the spectral reconstruction accuracy.

Spectral Super-Resolution for Multispectral Image Based on Spectral Improvement Strategy and Spatial Preservation Strategy

A spectral super-resolution method to recover a high-spectral-resolution HS image from multispectral (MS) images to address issues such as low spatial resolution, low temporal resolution, and some of the acquired spectral bands are either with low signal-to- noise ratio (SNR) or invalid because of the very high-noise level.

Deep Residual Network of Spectral and Spatial Fusion for Hyperspectral Image Super-Resolution

  • Xianhua HanYenwei Chen
  • Environmental Science, Mathematics
    2019 IEEE Fifth International Conference on Multimedia Big Data (BigMM)
  • 2019
Experimental results on benchmark datasets validate that the proposed residual CNN for hyperspectral super-resolution outperforms the state-of-the-art methods in both quantitative values and visual effect.

Spectral Superresolution of Multispectral Imagery With Joint Sparse and Low-Rank Learning

J-SLoL infers and recovers the unknown HS signals over a larger coverage by sparse coding on the learned dictionary pair, and validate the SSR performance on three HS–MS data sets in terms of reconstruction, classification, and unmixing by comparing with several existing state-of-the-art baselines, showing the effectiveness and superiority of the proposed J- SLoL algorithm.

Spatial and Spectral Joint Super-Resolution Using Convolutional Neural Network

Experimental results over simulated images from different sensors demonstrated that the proposed SepSSJSR1 is most effective to improve spatial and spectral resolution of MSIs sequentially by conducting spatial SR prior to spectral SR.

A Spectral–Spatial Jointed Spectral Super-Resolution and Its Application to HJ-1A Satellite Images

A spectral and spatial jointed spectral super-resolution method is proposed in this letter using an end-to-end learning strategy for each subspace with the cluster-based multibranch backpropagation neural network (BPNN).

Spectral Super-Resolution Based on Dictionary Optimization Learning via Spectral Library

The validity and superiority of the proposed SSR algorithm is confirmed by comparing it with several benchmark state-of-the-art approaches on different datasets by incorporating a spectral library as a priori in the spectral domain.

Residual HSRCNN: Residual Hyper-Spectral Reconstruction CNN from an RGB Image

A spectral reconstruction CNN for spectral super-resolution with an available RGB image, which predicts the high-frequency content of the fine spectral wavelength in narrow band interval is explored and a novel residual hyper-spectral reconstruction CNN framework is proposed to estimate the non-recovered high- frequencies from the output of the baseline CNN.

Separable-spectral convolution and inception network for hyperspectral image super-resolution

Experimental results show that adding several separable spectral convolutions and multi-path connection in a deep network can greatly improve the SR performance, and SSIN achieves higher accuracy and better visualization compare with other methods.

Hyperspectral image super-resolution combining with deep learning and spectral unmixing




Hyperspectral Super-Resolution by Coupled Spectral Unmixing

This paper proposes a method which performs hyperspectral super-resolution by jointly unmixing the two input images into the pure reflectance spectra of the observed materials and the associated mixing coefficients, with a number of useful constraints imposed by elementary physical properties of spectral mixing.

Sparse Spatio-spectral Representation for Hyperspectral Image Super-resolution

Comparison of the proposed approach with the state-of-the-art methods on both ground-based and remotely-sensed public hyperspectral image databases shows that the presented method achieves the lowest error rate on all test images in the three datasets.

Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization

Extensive experiments on image deblurring and super-resolution validate that by using adaptive sparse domain selection and adaptive regularization, the proposed method achieves much better results than many state-of-the-art algorithms in terms of both PSNR and visual perception.

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

This paper presents the first convolutional neural network capable of real-time SR of 1080p videos on a single K2 GPU and introduces an efficient sub-pixel convolution layer which learns an array of upscaling filters to upscale the final LR feature maps into the HR output.

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.

A Convex Formulation for Hyperspectral Image Superresolution via Subspace-Based Regularization

The split augmented Lagrangian shrinkage algorithm (SALSA), which is an instance of the alternating direction method of multipliers (ADMM), is added to this optimization problem, by means of a convenient variable splitting, and an effective algorithm is obtained that outperforms the state of the art.

Hierarchical Beta Process with Gaussian Process Prior for Hyperspectral Image Super Resolution

An image fusion based hyperspectral super resolution approach that employes a Bayesian representation model that accounts for spectral smoothness and spatial consistency of the representation by using Gaussian Processes and a spatial kernel in a hierarchical formulation of the Beta Process.

Generalized Assorted Pixel Camera: Postcapture Control of Resolution, Dynamic Range, and Spectrum

A comprehensive optimization method to arrive at the spatial and spectral layout of the color filter array of a GAP camera is presented and a novel algorithm for reconstructing the under-sampled channels of the image while minimizing aliasing artifacts is developed.

Learning a Deep Convolutional Network for Image Super-Resolution

This work proposes a deep learning method for single image super-resolution (SR) that directly learns an end-to-end mapping between the low/high-resolution images and shows that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network.

On Single Image Scale-Up Using Sparse-Representations

This paper deals with the single image scale-up problem using sparse-representation modeling, and assumes a local Sparse-Land model on image patches, serving as regularization, to recover an original image from its blurred and down-scaled noisy version.