Cascaded Convolutional Neural Network-Based Hyperspectral Image Resolution Enhancement via an Auxiliary Panchromatic Image

LRTCFPan: Low-Rank Tensor Completion Based Framework for Pansharpening

Pansharpening refers to the fusion of a low spatial-resolution multispectral image with a high spatial-resolution panchromatic image. In this paper, we propose a novel low-rank tensor completion (LRTC)-based framework with some regularizers for multispectral image pansharpening, called LRTCFPan. The tensor completion technique is commonly used for image recovery, but it cannot directly perform the pansharpening or, more generally, the super-resolution problem because of the formulation gap. Different from previous variational methods, we first formulate a pioneering image super-resolution (ISR) degradation model, which equivalently removes the downsampling operator and transforms the tensor completion framework. Under such a framework, the original pansharpening problem is realized by the LRTC-based technique with some deblurring regularizers. From the perspective of regularizer, we further explore a local-similarity-based dynamic detail mapping (DDM) term to more accurately capture the spatial content of the panchromatic image. Moreover, the low-tubal-rank property of multispectral images is investigated, and the low-tubal-rank prior is introduced for better completion and global characterization. To solve the proposed LRTCFPan model, we develop an alternating direction method of multipliers (ADMM)-based algorithm. Comprehensive experiments at reduced-resolution (i.e., simulated) and full-resolution (i.e., real) data exhibit that the LRTCFPan method significantly outperforms other state-of-the-art pansharpening methods. The code is publicly available at: https://github.com/zhongchengwu/code_LRTCFPan.

Spatial response resampling (SR2): Accounting for the spatial point spread function in hyperspectral image resampling

With the increased availability of hyperspectral imaging (HSI) data at various scales (0.03-30 m), the role of simulation is becoming increasingly important in data analysis and applications. There are few commercially available tools to spatially degrade imagery based on the spatial response of a coarser resolution sensor. Instead, HSI data are typically spatially degraded using nearest neighbor, pixel aggregate or cubic convolution approaches. Without accounting for the spatial response of the simulated sensor, these approaches yield unrealistically sharp images. This article describes the spatial response resampling (SR²) workflow, a novel approach to degrade georeferenced raster HSI data based on the spatial response of a coarser resolution sensor. The workflow is open source and widely available for personal, academic or commercial use with no restrictions. The importance of the SR² workflow is shown with three practical applications (data cross-validation, flight planning and data fusion of separate VNIR and SWIR images).•The SR² workflow derives the point spread function of a specified HSI sensor based on nominal data acquisition parameters (e.g., integration time, altitude, speed), convolving it with a finer resolution HSI dataset for data simulation.•To make the workflow approachable for end users, we provide a MATLAB function that implements the SR² methodology.

Hyperspectral Image Super-Resolution Based on Spatial Correlation-Regularized Unmixing Convolutional Neural Network

Super-resolution (SR) technology has emerged as an effective tool for image analysis and interpretation. However, single hyperspectral (HS) image SR remains challenging, due to the high spectral dimensionality and lack of available high-resolution information of auxiliary sources. To fully exploit the spectral and spatial characteristics, in this paper, a novel single HS image SR approach is proposed based on a spatial correlation-regularized unmixing convolutional neural network (CNN). The proposed approach takes advantage of a CNN to explore the collaborative spatial and spectral information of an HS image and infer the high-resolution abundance maps, thereby reconstructing the anticipated high-resolution HS image via the linear spectral mixture model. Moreover, a dual-branch architecture network and spatial spread transform function are employed to characterize the spatial correlation between the high- and low-resolution HS images, aiming at promoting the fidelity of the super-resolved image. Experiments on three public remote sensing HS images demonstrate the feasibility and superiority in terms of spectral fidelity, compared with some state-of-the-art HS image super-resolution methods.