A Fast Multiscale Spatial Regularization for Sparse Hyperspectral Unmixing

IEEE Trans. Comput. Imaging

2020

Self Cite

Endmember (EM) spectral variability can greatly impact the performance of standard hyperspectral image analysis algorithms. Extended parametric models have been successfully applied to account for the EM spectral variability. However, these models still lack the compromise between flexibility and low-dimensional representation that is necessary to properly explore the fact that spectral variability is often confined to a low-dimensional manifold in real scenes. In this paper we propose to learn a spectral variability model directly from the observed data, instead of imposing it a priori. This is achieved through a deep generative EM model, which is estimated using a variational autoencoder (VAE). The encoder and decoder that compose the generative model are trained using pure pixel information extracted directly from the observed image, what allows for an unsupervised formulation. The proposed EM model is applied to the solution of a spectral unmixing problem, which we cast as an alternating nonlinear least-squares problem that is solved iteratively with respect to the abundances and to the low-dimensional representations of the EMs in the latent space of the deep generative model. Simulations using both synthetic and real data indicate that the proposed strategy can outperform the competing state-of-the-art algorithms.

Section: Real Datamentioning

confidence: 99%

Deep Generative Endmember Modeling: An Application to Unsupervised Spectral Unmixing

IEEE Trans. Comput. Imaging

2020

Self Cite

“…Specifically, W must group pixels that are spatially adjacent and spectrally similar and must respect image borders by not grouping pixels corresponding to different image structures or features. Following the same approach as in [33], [34], we consider the superpixel decomposition of the image Y for the transformation W . Besides satisfying the criteria outlined above, multiscale decompositions based on superpixel algorithms have shown excellent performance in SU considering both sparsity [33] and variability of the endmembers [34].…”

Section: A Unmixing In the Coarse Scalementioning

confidence: 99%

“…In this paper, we propose a new multiscale spatial regularization approach for kernel-based nonlinear unmixing. Building upon the ideas proposed in [33], we employ a multiscale representation to divide the unmixing problem into two simpler problems in different scales. Though based on the same principle used in [33], devising kernel-based mixing models in multiple scales is more challenging than in the linear case.…”

Section: Introductionmentioning

confidence: 99%

“…Building upon the ideas proposed in [33], we employ a multiscale representation to divide the unmixing problem into two simpler problems in different scales. Though based on the same principle used in [33], devising kernel-based mixing models in multiple scales is more challenging than in the linear case. Moreover, we address the parameter adjustment problem differently from what has been done in previous multiscale SU formulations in [33], [34].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Blind Multiscale Spatial Regularization Framework for Kernel-Based Spectral Unmixing

IEEE Trans. on Image Process.

et al. 2020

Self Cite

Introducing spatial prior information in hyperspectral imaging (HSI) analysis has led to an overall improvement of the performance of many HSI methods applied for denoising, classification, and unmixing. Extending such methodologies to nonlinear settings is not always straightforward, specially for unmixing problems where the consideration of spatial relationships between neighboring pixels might comprise intricate interactions between their fractional abundances and nonlinear contributions. In this paper, we consider a multiscale regularization strategy for nonlinear spectral unmixing with kernels. The proposed methodology splits the unmixing problem into two sub-problems at two different spatial scales: a coarse scale containing lowdimensional structures, and the original fine scale. The coarse spatial domain is defined using superpixels that result from a multiscale transformation. Spectral unmixing is then formulated as the solution of quadratically constrained optimization problems, which are solved efficiently by exploring a reformulation of their dual cost functions in the form of root-finding problems. Furthermore, we employ a theory-based statistical framework to devise a consistent strategy to estimate all required parameters, including both the regularization parameters of the algorithm and the number of superpixels of the transformation, resulting in a truly blind (from the parameters setting perspective) unmixing method. Experimental results attest the superior performance of the proposed method when comparing with other, state-of-theart, related strategies.

“…More recently, subspace-based formulations have become the leading approach to solve this problem, exploring the natural low-dimensional representation of HSIs as a linear combination of a small set of basis vectors or spectral signatures [4], [9], [10], [11]. Many approaches have been proposed to perform image fusion under this formulation.…”

Section: Introductionmentioning

confidence: 99%

Super-Resolution for Hyperspectral and Multispectral Image Fusion Accounting for Seasonal Spectral Variability

IEEE Trans. on Image Process.

2020

Self Cite

100

Image fusion combines data from different heterogeneous sources to obtain more precise information about an underlying scene. Hyperspectral-multispectral (HS-MS) image fusion is currently attracting great interest in remote sensing since it allows the generation of high spatial resolution HS images, circumventing the main limitation of this imaging modality. Existing HS-MS fusion algorithms, however, neglect the spectral variability often existing between images acquired at different time instants. This time difference causes variations in spectral signatures of the underlying constituent materials due to different acquisition and seasonal conditions. This paper introduces a novel HS-MS image fusion strategy that combines an unmixingbased formulation with an explicit parametric model for typical spectral variability between the two images. Simulations with synthetic and real data show that the proposed strategy leads to a significant performance improvement under spectral variability and state-of-the-art performance otherwise.