A Gaussian Mixture Model Representation of Endmember Variability in Hyperspectral Unmixing

  title={A Gaussian Mixture Model Representation of Endmember Variability in Hyperspectral Unmixing},
  author={Yuan Zhou and Anand Rangarajan and Paul D. Gader},
  journal={IEEE Transactions on Image Processing},
Hyperspectral unmixing while considering endmember variability is usually performed by the normal compositional model, where the endmembers for each pixel are assumed to be sampled from unimodal Gaussian distributions. However, in real applications, the distribution of a material is often not Gaussian. In this paper, we use Gaussian mixture models (GMM) to represent endmember variability. We show, given the GMM starting premise, that the distribution of the mixed pixel (under the linear mixing… 

Hyperspectral Unmixing with Gaussian Mixture Model and Spatial Group Sparsity

To achieve better abundance estimation performance, a method based on the Gaussian mixture model (GMM) and spatial group sparsity constraint is proposed, which has higher unmixing precision compared with other state-of-the-art methods.

Hyperspectral Unmixing with Gaussian Mixture Model and Low-Rank Representation

This work proposes a novel GMM unmixing method based on superpixel segmentation (SS) and low-rank representation (LRR), which is called GMM-SS-LRR, which is efficient compared with other current popular methods.

Bayesian Unmixing of Hyperspectral Image Sequence With Composite Priors for Abundance and Endmember Variability

A Bayesian unmixing model considering spectral variability for hyperspectral sequence is proposed, in which composite prior distributions of abundance and endmember variability are developed.

Gaussian Mixture Model for Hyperspectral Unmixing with Low-Rank Representation

  • Qiwen JinYong Ma Jun Huang
  • Environmental Science, Mathematics
    IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium
  • 2019
A new unmixing method with superpixel segmentation and low-rank representation based on GMM called GMM-SS-LRR, which can consider the local spatial correlation of HSI and is efficient compared with other current popular methods.

Augmented Gaussian Linear Mixture Model for Spectral Variability in Hyperspectral Unmixing

This paper reformulates the LMM by adding a term to account for the spectral variations of endmember spectra of the dictionary, and employs the multiplicative updating rules to accelerate the convergence and exploit the sparsity of the unknown parameters.

Spectral Variability Aware Blind Hyperspectral Image Unmixing Based on Convex Geometry

An integrated unmixing chain is proposed which tries to adress the shortcomings of the classical tools used in the linear case, based on the previously proposed extended linear mixing model, and shows the interest of the proposed approach on simulated and real datasets.

Variational Autoencoders for Hyperspectral Unmixing with Endmember Variability

This paper presents a variational autoencoder (VAE) framework for hyperspectral unmixing accounting for the endmember variability that is able to fit an arbitrary distribution of endmembers for each material through the representation capacity of deep neural networks.

Hierarchical Sparse Nonnegative Matrix Factorization for Hyperspectral Unmixing with Spectral Variability

The proposed unmixing model not only extracts and clusters the prototypal endmember spectra, but also estimates the abundances of each endmember.

Robust Supervised Method for Nonlinear Spectral Unmixing Accounting for Endmember Variability

This work developed data sets of homogeneously mixed mineral powder mixtures, acquired by two different sensors, an Agrispec spectrometer and a snapscan shortwave infrared (SWIR) hyperspectral camera, under strictly controlled experimental settings.



A Spatial Compositional Model for Linear Unmixing and Endmember Uncertainty Estimation

The spatial compositional model (SCM) is derived from the ground up without the pixel independence assumption, resulting in an algorithm that estimates endmembers, abundances, noise variances, and endmember uncertainty simultaneously.

Hyperspectral Unmixing in Presence of Endmember Variability, Nonlinearity, or Mismodeling Effects

The proposed mixture and Bayesian models and their estimation algorithms are validated on both synthetic and real images showing competitive results regarding the quality of the inferences and the computational complexity, when compared with the state-of-the-art algorithms.

Unsupervised Unmixing of Hyperspectral Images Accounting for Endmember Variability

This paper presents an unsupervised Bayesian algorithm for hyperspectral image unmixing, accounting for endmember variability, and proposes to use a Hamiltonian Monte Carlo algorithm.

Bayesian Estimation of Linear Mixtures Using the Normal Compositional Model. Application to Hyperspectral Imagery

This paper proposes to estimate the mixture coefficients (referred to as abundances) using a Bayesian algorithm and builds a hybrid Gibbs sampler to generate abundance and variance samples distributed according to the joint posterior of the abundances and noise variances.

Blind Hyperspectral Unmixing Using an Extended Linear Mixing Model to Address Spectral Variability

The proposed algorithm to unmix hyperspectral data using a recently proposed extended LMM outperforms other methods aimed at addressing spectral variability, and can provide an accurate estimation of endmember variability along the scene because of the scaling factors estimation.

Hyperspectral Unmixing With Spectral Variability Using a Perturbed Linear Mixing Model

A new linear mixing model is introduced that explicitly accounts for spatial and spectral endmember variabilities and can be estimated using an optimization algorithm based on the alternating direction method of multipliers.

Estimating the Number of Endmembers in Hyperspectral Images Using the Normal Compositional Model and a Hierarchical Bayesian Algorithm

This paper proposes to estimate the mixture coefficients of the Normal Compositional Model (referred to as abundances) as well as their number using a reversible jump Bayesian algorithm.

Nonlinear Unmixing of Hyperspectral Images Using a Generalized Bilinear Model

A generalized bilinear model and a hierarchical Bayesian algorithm for unmixing hyperspectral images and a Metropolis-within-Gibbs algorithm is proposed, which allows samples distributed according to this posterior to be generated and to estimate the unknown model parameters.

A Review of Nonlinear Hyperspectral Unmixing Methods

This paper aims to give an historical overview of the majority of nonlinear mixing models and nonlinear unmixing methods, and to explain some of the more popular techniques in detail.

PCE: Piecewise Convex Endmember Detection

  • Alina ZareP. Gader
  • Environmental Science, Mathematics
    IEEE Transactions on Geoscience and Remote Sensing
  • 2010
Algorithm results on hyperspectral data indicate that PCE produces endmembers that represent the true ground-truth classes of the input data set, thus incorporating an endmember's spectral variability.