New Bayesian approach for semi-supervised hyperspectral unmixing in linear mixing models

Amiri, Fahime; Kahaei, Mohammad Hossein

doi:10.1109/iraniancee.2017.7985334

Cited by 5 publications

(4 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we propose the Sparse Dirichlet Prior with PPNMM (SDP-PPNMM) algorithm in a semi-supervised manner that means we do not need any EEA and the lack of knowledge about pure endmembers is compensated just by selecting suitable priors. In fact, we extend our previous work [4] to a nonlinear mixing model. We assume that a large library of endmembers is available, which is a realistic assumption due to collecting a wide variety of spectral signature of various common materials during few decades [5].…”

Section: Introductionmentioning

confidence: 53%

“…In this paper, we propose the sparse Dirichlet prior with the PPNMM (SDP-PPNMM) algorithm in a semisupervised manner in which we compensate the lack of knowledge about pure endmembers by selecting suitable priors 1 that model sparse abundance vectors. In this way, we extend our previous work [4] to a nonlinear mixing model. We assume that a large library of endmembers is available since a wide variety of spectral signatures of various materials have already been collected during the last few decades [5].…”

Section: Introductionmentioning

confidence: 64%

See 1 more Smart Citation

Bayesian unmixing using sparse dirichlet prior with polynomial post‐nonlinear mixing model

Amiri

Kahaei

2018

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

A sparse Dirichlet prior is proposed for estimating the abundance vector of hyperspectral images with a nonlinear mixing model. This sparse prior leads to an unmixing procedure in a semisupervised scenario in which exact materials are unknown. The nonlinear model is a polynomial post-nonlinear mixing model, which represents each hyperspectral pixel as a nonlinear function of pure spectral signatures corrupted by additive white noise. Simulation results show more than 50% reduction in the estimation error.

show abstract

Section: Introductionmentioning

confidence: 53%

Section: Introductionmentioning

confidence: 64%

Bayesian unmixing using sparse dirichlet prior with polynomial post‐nonlinear mixing model

Amiri

Kahaei

2018

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

show abstract

“…In [15] the authors propose a two-level hierarchical prior equivalent to Laplace but maintaining the conjugacy for the abundances to promote sparsity. In [16], [17], similar Dirichlet model priors show the possibility of promoting sparsity among the abundances respectively.…”

Section: Introductionmentioning

confidence: 99%

Sparse Spectral Unmixing of Hyperspectral Images using Expectation-Propagation

Zeng

Altmann

Chen

et al. 2020

2020 IEEE International Conference on Visual Communications and Image Processing (VCIP)

View full text Add to dashboard Cite

This paper presents a novel Bayesian approach for hyperspectral image unmixing. The observed pixels are modeled by a linear combination of material signatures weighted by their corresponding abundances. A spike-and-slab abundance prior is adopted to promote sparse mixtures and an Ising prior model is used to capture spatial correlation of the mixture support across pixels. We approximate the posterior distribution of the abundances using the expectation-propagation (EP) method. We show that it can significantly reduce the computational complexity of the unmixing stage and meanwhile provide uncertainty measures, compared to expensive Monte Carlo strategies traditionally considered for uncertainty quantification. Moreover, many variational parameters within each EP factor can be updated in a parallel manner, which enables mapping of efficient algorithmic architectures based on graphics processing units (GPU). Under the same approximate Bayesian framework, we then extend the proposed algorithm to semi-supervised unmixing, whereby the abundances are viewed as latent variables and the expectation-maximization (EM) algorithm is used to refine the endmember matrix. Experimental results on synthetic data and real hyperspectral data illustrate the benefits of the proposed framework over state-of-art linear unmixing methods.

show abstract

“…Most scholars initially used traditional processing methods, such as the support vector machine (SVM) [ 5 ], k nearest neighbor classification algorithm (KNN) [ 6 ], and the Bayesian network [ 7 ], for HSI to classify surficial objects. However, these classification results were not ideal.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Remote Sensing Image Classification Based on Maximum Overlap Pooling Convolutional Neural Network

Yang

et al. 2018

Sensors

View full text Add to dashboard Cite

In a traditional convolutional neural network structure, pooling layers generally use an average pooling method: a non-overlapping pooling. However, this condition results in similarities in the extracted image features, especially for the hyperspectral images of a continuous spectrum, which makes it more difficult to extract image features with differences, and image detail features are easily lost. This result seriously affects the accuracy of image classification. Thus, a new overlapping pooling method is proposed, where maximum pooling is used in an improved convolutional neural network to avoid the fuzziness of average pooling. The step size used is smaller than the size of the pooling kernel to achieve overlapping and coverage between the outputs of the pooling layer. The dataset selected for this experiment was the Indian Pines dataset, collected by the airborne visible/infrared imaging spectrometer (AVIRIS) sensor. Experimental results show that using the improved convolutional neural network for remote sensing image classification can effectively improve the details of the image and obtain a high classification accuracy.

show abstract

New Bayesian approach for semi-supervised hyperspectral unmixing in linear mixing models

Cited by 5 publications

References 43 publications

Bayesian unmixing using sparse dirichlet prior with polynomial post‐nonlinear mixing model

Bayesian unmixing using sparse dirichlet prior with polynomial post‐nonlinear mixing model

Sparse Spectral Unmixing of Hyperspectral Images using Expectation-Propagation

Hyperspectral Remote Sensing Image Classification Based on Maximum Overlap Pooling Convolutional Neural Network

Contact Info

Product

Resources

About