Linear unmixing protocol for hyperspectral image fusion analysis applied to a case study of vegetal tissues

Hyperspectral imaging (HSI) is a useful non-invasive technique that offers spatial and chemical information of samples. Often, different HSI techniques are used to obtain complementary information from the sample by combining different image modalities (Image Fusion). However, issues related to the different spatial resolution, sample orientation or area scanned among platforms need to be properly addressed. Unmixing methods are helpful to analyze and interpret the information of HSI related to each of the components contributing to the signal. Among those, Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) offers very suitable features for image fusion, since it can easily cope with multiset structures formed by blocks of images coming from different samples and platforms and allows the use of optional and diverse constraints to adapt to the specific features of each HSI employed. In this work, a case study based on the investigation of cross-sections from rice leaves by Raman, synchrotron infrared and fluorescence imaging techniques is presented. HSI of these three different techniques are fused for the first time in a single data structure and analyzed by MCR-ALS. This example is challenging in nature and is particularly suitable to describe clearly the necessary steps required to perform unmixing in an image fusion context. Although this protocol is presented and applied to a study of vegetal tissues, it can be generally used in many other samples and combinations of imaging platforms.

Fast and Stable Hyperspectral Multispectral Image Fusion Technique Using Moore–Penrose Inverse Solver

Fusion low-resolution hyperspectral images (LR-HSI) and high-resolution multispectral images (HR-MSI) are important methods for obtaining high-resolution hyperspectral images (HR-HSI). Some hyperspectral image fusion application areas have strong real-time requirements for image fusion, and a fast fusion method is urgently needed. This paper proposes a fast and stable fusion method (FSF) based on matrix factorization, which can largely reduce the computational workloads of image fusion to achieve fast and efficient image fusion. FSF introduces the Moore–Penrose inverse in the fusion model to simplify the estimation of the coefficient matrix and uses singular value decomposition (SVD) to simplify the estimation of the spectral basis, thus significantly reducing the computational effort of model solving. Meanwhile, FSF introduces two multiplicative iterative processes to optimize the spectral basis and coefficient matrix to achieve stable and high-quality fusion. We have tested the fusion method on remote sensing and ground-based datasets. The experiments show that our proposed method can achieve the performance of several state-of-the-art algorithms while reducing execution time to less than 1% of such algorithms.

Variational Fusion of Hyperspectral Data by Non-Local Filtering

The fusion of multisensor data has attracted a lot of attention in computer vision, particularly among the remote sensing community. Hyperspectral image fusion consists in merging the spectral information of a hyperspectral image with the geometry of a multispectral one in order to infer an image with high spatial and spectral resolutions. In this paper, we propose a variational fusion model with a nonlocal regularization term that encodes patch-based filtering conditioned to the geometry of the multispectral data. We further incorporate a radiometric constraint that injects the high frequencies of the scene into the fused product with a band per band modulation according to the energy levels of the multispectral and hyperspectral images. The proposed approach proved robust to noise and aliasing. The experimental results demonstrate the performance of our method with respect to the state-of-the-art techniques on data acquired by commercial hyperspectral cameras and Earth observation satellites.