Nonnegative sparse autoencoder for robust endmember extraction from remotely sensed hyperspectral images

Su, Yuanchao; Marinoni, Andrea; Li, Jun; Gamba, Paolo

doi:10.1109/igarss.2017.8126930

Cited by 26 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This leads to vanishing gradients during training. The methods in [19,82] use a parameterized sigmoid activation function. The ReLU activation largely mitigates the problem of vanishing gradients, but the function is still saturated to the left, and hidden units can become stuck at zero.…”

Section: B Choice Of Activation Function For Hidden Layersmentioning

confidence: 99%

See 1 more Smart Citation

Blind Hyperspectral Unmixing Using Autoencoders: A Critical Comparison

Palsson

Sveinsson

Ulfarsson

2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Section: B Choice Of Activation Function For Hidden Layersmentioning

confidence: 99%

“…All the methods in [19,71,78,82,84,88,95,96] use the MSE measure as the fidelity term of the network's loss function. It is also possible to use a combination of both scale-invariant and non-scale-invariant loss terms.…”

Section: F Choices Of Loss Fidelity Function and Spectral Variabilitymentioning

confidence: 99%

Blind Hyperspectral Unmixing Using Autoencoders: A Critical Comparison

Palsson

Sveinsson

Ulfarsson

2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

“…Recent applications of neural network methods for unmixing are [23], [24], and [25] where an autoencoder is used for abundance estimation, i.e., mapping input spectra to abundance fractions but not extracting the actual endmembers. In [26], a shallow symmetric, i.e., the encoder and decoder have tied weights, nonnegative sparse autoencoder is used to extract endmembers. The novelty of this method lies in the use of an automatic sampler with a local outlier factor and affinity propagation for intelligently selecting samples for the training set.…”

Section: Introductionmentioning

confidence: 99%

“…Almost all other deep learning based methods for HSU do not perform blind unmixing, i.e., estimate both the endmember spectra and their abundances. To the authors' best knowledge, the only deep learning methods that perform blind unmixing, are the methods in [26]- [29]. The proposed method differs mainly from these methods by having both a deep encoder and being able to exploit the sparsity of abundances through a layer using a custom activation function instead of explicit sparsity regularization.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Unmixing Using a Neural Network Autoencoder

et al. 2018

View full text Add to dashboard Cite

In this paper, we present a deep learning based method for blind hyperspectral unmixing in the form of a neural network autoencoder. We show that the linear mixture model implicitly puts certain architectural constraints on the network, and it effectively performs blind hyperspectral unmixing. Several different architectural configurations of both shallow and deep encoders are evaluated. Also, deep encoders are tested using different activation functions. Furthermore, we investigate the performance of the method using three different objective functions. The proposed method is compared to other benchmark methods using real data and previously established ground truths of several common data sets. Experiments show that the proposed method compares favorably to other commonly used hyperspectral unmixing methods and exhibits robustness to noise. This is especially true when using spectral angle distance as the network's objective function. Finally, results indicate that a deeper and a more sophisticated encoder does not necessarily give better results.

show abstract

“…To tackle nonlinearities, in [16], nonlinear kernelized NMF was presented. More recently, unmixing based on autoencoders [17][18][19][20][21][22] are employed to estimate endmembers and fractional abundances simultaneously. However, these algorithms are either limited to the linear mixing model or implementation of an existing nonlinear (bilinear) model to the autoencoder framework.…”

Section: Introductionmentioning

confidence: 99%

A Supervised Method for Nonlinear Hyperspectral Unmixing

et al. 2019

View full text Add to dashboard Cite

Due to the complex interaction of light with the Earth’s surface, reflectance spectra can be described as highly nonlinear mixtures of the reflectances of the material constituents occurring in a given resolution cell of hyperspectral data. Our aim is to estimate the fractional abundance maps of the materials from the nonlinear hyperspectral data. The main disadvantage of using nonlinear mixing models is that the model parameters are not properly interpretable in terms of fractional abundances. Moreover, not all spectra of a hyperspectral dataset necessarily follow the same particular mixing model. In this work, we present a supervised method for nonlinear spectral unmixing. The method learns a mapping from a true hyperspectral dataset to corresponding linear spectra, composed of the same fractional abundances. A simple linear unmixing then reveals the fractional abundances. To learn this mapping, ground truth information is required, in the form of actual spectra and corresponding fractional abundances, along with spectra of the pure materials, obtained from a spectral library or available in the dataset. Three methods are presented for learning nonlinear mapping, based on Gaussian processes, kernel ridge regression, and feedforward neural networks. Experimental results conducted on an artificial dataset, a data set obtained by ray tracing, and a drill core hyperspectral dataset shows that this novel methodology is very promising.

show abstract

Nonnegative sparse autoencoder for robust endmember extraction from remotely sensed hyperspectral images

Cited by 26 publications

References 11 publications

Blind Hyperspectral Unmixing Using Autoencoders: A Critical Comparison

Blind Hyperspectral Unmixing Using Autoencoders: A Critical Comparison

Hyperspectral Unmixing Using a Neural Network Autoencoder

A Supervised Method for Nonlinear Hyperspectral Unmixing

Contact Info

Product

Resources

About