Stacked Nonnegative Sparse Autoencoders for Robust Hyperspectral Unmixing

Su, Yuanchao; Marinoni, Andrea; Zhang, Liangpei; Plaza, Javier; Gamba, Paolo

doi:10.1109/lgrs.2018.2841400

Cited by 100 publications

(45 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three of the comparison methods are based on deep learning. The comparison methods are vertex component analysis (VCA) [11], sparsity constrained nonnegative matrix factorization ( 1/2 -NMF) [13], sticky hierarchical Dirichlet process (SHDP) [18] which is a spatial-spectral blind unmixing method, spatial group sparsity regularized nonnegative matrix factorization (SGSRNMF) which is a spatial-spectral blind unmixing method [42], deep autoencoder unmixing (DAEU), an autoencoder based method described in [6], stacked nonnegative sparse autoencoder (SNSA) unmixing method described in [47], and an untied denoising autoencoder with sparsity (uDAS) unmixing method described in [48].…”

Section: Methodsmentioning

confidence: 99%

Spectral-Spatial Hyperspectral Unmixing Using Multitask Learning

2019

View full text Add to dashboard Cite

Hyperspectral unmixing is an important and challenging task in the field of remote sensing which arises when the spatial resolution of sensors is insufficient for the separation of spectrally distinct materials. Hyperspectral images, like other natural images, have highly correlated pixels and it is very desirable to make use of this spatial information. In this paper, a deep learning based method for blind hyperspectral unmixing is presented. The method uses multitask learning through multiple parallel autoencoders to unmix a neighborhood of pixels simultaneously. Operating on image patches instead of single pixels enables the method to take advantage of spatial information in the hyperspectral image. The method is the first in its class to directly utilize the spatial structure of hyperspectral images (HSIs) for the estimation of the spectral signatures of endmembers in the data cube. We evaluate the proposed method using two real HSIs and compare it to seven state-of-the-art methods that either rely only on spectral or both on spectral and spatial information in the HSIs. The proposed method outperforms all the baseline unmixing methods in experiments. Hyperspectral, unmixing, autoencoder, multitask learning, deep learning. INDEX TERMS

show abstract

Section: Methodsmentioning

confidence: 99%

Spectral-Spatial Hyperspectral Unmixing Using Multitask Learning

2019

View full text Add to dashboard Cite

show abstract

“…In addition, we also provide the test results obtained with NMF-based methods, NMF-sp [45] and collaborative nonnegative matrix factorization (CNMF) [46]) and topic-based methods, PLSA [33], PLSA-sp [22], which introduces a sparsity constraint over the documents, and LDA [32], as shown in the latest papers [23]. To further illustrate the effectiveness of the proposed method for complex datasets, the results of some recent methods, a deep autoencoder network (DAEN) [16], a stacked nonnegative sparse autoencoder (SNSA) [18], a deep autoencoder unmixing (DAEU) [47], and dyadic cyclic descent optimization (DCD) [48], are shown in the results of the Jasper and Urban datasets [16]- [19].…”

Section: A Experimental Settingsmentioning

confidence: 99%

A Sparse Topic Relaxion and Group Clustering Model for Hyperspectral Unmixing

Zhu

Wang

Zeng

et al. 2021

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

View full text Add to dashboard Cite

Hyperspectral unmixing has been a hot research topic in the field of hyperspectral remote sensing. In recent years, the employment of the probabilistic topic model to acquire the latent topics of hyperspectral images has been an effective method for spectral unmixing. However, such methods fail to fully exploit the potential of topic models in uncovering image semantics, and they need extra sparsity constraints, which greatly increases the complexity of the model. To solve these problems, a sparse topic relaxion and group clustering model for hyperspectral unmixing (STRGC) is proposed. In STRGC, the sparse prior constraints implied by the sparse topic model are introduced, which means that the sparse characteristics of the model are used to capture the semantic representation of the spectrum. Through the relaxation of the model, the possible spectral representations of ground features can be obtained, and this further alleviates the influence caused by endmember variability on the accuracy of the unmixing process. Then, fuzzy clustering is used to locate the position of the endmember quickly and accurately. Furthermore, unmixing models with different characteristics are united to alleviate the ill-posed nature of the model, thereby improving the fractional abundance. Experimental results obtained with one simulated dataset and three well-known real hyperspectral datasets confirm the effectiveness and advantages of the proposed method.

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

Stacked Nonnegative Sparse Autoencoders for Robust Hyperspectral Unmixing

Cited by 100 publications

References 12 publications

Spectral-Spatial Hyperspectral Unmixing Using Multitask Learning

Spectral-Spatial Hyperspectral Unmixing Using Multitask Learning

A Sparse Topic Relaxion and Group Clustering Model for Hyperspectral Unmixing

A Supervised Method for Nonlinear Hyperspectral Unmixing

Contact Info

Product

Resources

About