Abstract:Abstract-Hyperspectral images contain mixed pixels due to low spatial resolution of hyperspectral sensors. Spectral unmixing problem refers to decomposing mixed pixels into a set of endmembers and abundance fractions. Due to nonnegativity constraint on abundance fractions, nonnegative matrix factorization (NMF) methods have been widely used for solving spectral unmixing problem. In this letter we proposed using multilayer NMF (MLNMF) for the purpose of hyperspectral unmixing. In this approach, spectral signatu… Show more
“…To make a holistic comparison, five state-of-the-art endmember extraction methods including ICA [50], MLNMF [29], CONMF [51], RNMF [30] and AA [36] have been implemented and their endmember results are compared against that of PWAA-EMD. The extracted endmembers are evaluated with two popular measures, spectral angle distance (SAD) and root-mean-square-error (RMSE).…”
Section: Resultsmentioning
confidence: 99%
“…The matrix factorization methods formulate an optimization problem of blind source decomposition with many additive constraints (e.g., sparsity, low rank or positivity) to simultaneously estimate all the endmembers. Representative algorithms include sparse nonnegative matrix underapproximation (SNMU) [28], multilayer nonnegative matrix factorization (MLNMF) [29], robust nonnegative matrix factorization (RNMF) [30] and constrained nonnegative matrix factorization [12]. The statistical methods transform endmember extraction into a statistical inference problem and aim for the highly mixed hyperspectral image scenarios, with typical examples of independent component analysis (ICA) [31] and Bayesian algorithms such as normal endmember spectral unmixing [32] and the hierarchical Bayesian algorithm [33].…”
Abstract:A Probabilistic Weighted Archetypal Analysis method with Earth Mover's Distance (PWAA-EMD) is proposed to extract endmembers from hyperspectral imagery (HSI). The PWAA-EMD first utilizes the EMD dissimilarity matrix to weight the coefficient matrix in the regular Archetypal Analysis (AA). The EMD metric considers manifold structures of spectral signatures in the HSI data and could better quantify the dissimilarity features among pairwise pixels. Second, the PWAA-EMD adopts the Bayesian framework and formulates the improved AA into a probabilistic inference problem by maximizing a joint posterior density. Third, the optimization problem is solved by the iterative multiplicative update scheme, with a careful initialization from the two-stage algorithm and the proper endmembers are finally obtained. The synthetic and real Cuprite Hyperspectral datasets are utilized to verify the performance of PWAA-EMD and five popular methods are implemented to make comparisons. The results show that PWAA-EMD surpasses all the five methods in the average results of spectral angle distance (SAD) and root-mean-square-error (RMSE). Especially, the PWAA-EMD obtains more accurate estimation than AA in almost all the classes of endmembers including two similar ones. Therefore, the PWAA-EMD could be an alternative choice for endmember extraction on the hyperspectral data.
“…To make a holistic comparison, five state-of-the-art endmember extraction methods including ICA [50], MLNMF [29], CONMF [51], RNMF [30] and AA [36] have been implemented and their endmember results are compared against that of PWAA-EMD. The extracted endmembers are evaluated with two popular measures, spectral angle distance (SAD) and root-mean-square-error (RMSE).…”
Section: Resultsmentioning
confidence: 99%
“…The matrix factorization methods formulate an optimization problem of blind source decomposition with many additive constraints (e.g., sparsity, low rank or positivity) to simultaneously estimate all the endmembers. Representative algorithms include sparse nonnegative matrix underapproximation (SNMU) [28], multilayer nonnegative matrix factorization (MLNMF) [29], robust nonnegative matrix factorization (RNMF) [30] and constrained nonnegative matrix factorization [12]. The statistical methods transform endmember extraction into a statistical inference problem and aim for the highly mixed hyperspectral image scenarios, with typical examples of independent component analysis (ICA) [31] and Bayesian algorithms such as normal endmember spectral unmixing [32] and the hierarchical Bayesian algorithm [33].…”
Abstract:A Probabilistic Weighted Archetypal Analysis method with Earth Mover's Distance (PWAA-EMD) is proposed to extract endmembers from hyperspectral imagery (HSI). The PWAA-EMD first utilizes the EMD dissimilarity matrix to weight the coefficient matrix in the regular Archetypal Analysis (AA). The EMD metric considers manifold structures of spectral signatures in the HSI data and could better quantify the dissimilarity features among pairwise pixels. Second, the PWAA-EMD adopts the Bayesian framework and formulates the improved AA into a probabilistic inference problem by maximizing a joint posterior density. Third, the optimization problem is solved by the iterative multiplicative update scheme, with a careful initialization from the two-stage algorithm and the proper endmembers are finally obtained. The synthetic and real Cuprite Hyperspectral datasets are utilized to verify the performance of PWAA-EMD and five popular methods are implemented to make comparisons. The results show that PWAA-EMD surpasses all the five methods in the average results of spectral angle distance (SAD) and root-mean-square-error (RMSE). Especially, the PWAA-EMD obtains more accurate estimation than AA in almost all the classes of endmembers including two similar ones. Therefore, the PWAA-EMD could be an alternative choice for endmember extraction on the hyperspectral data.
“…In this paper, we have considered the most representative MVA approaches [4] such as principal component analysis (PCA) [2], canonical correlation analysis (CCA) [2], nonnegative matrix factorization [11], entropy component analysis, and also independent component analysis as dimensionality reduction methods. These MVA approaches include dimensionality reduction, feature extraction, and feature selection techniques and they are widely used in processing high-dimensional data.…”
This paper contributes the concept of spectralspatial kernel-based multivariate analysis (KMVSSA) based on the statistical principle of multivariate statistics. The essence of proposed framework is to expose the inherent structure and meaning revealed within spectral and spatial features through various statistical methods in hyperspectral remotely sensed data. This kernel-based framework is investigated to incorporate the spectral and spatial information simultaneously for dimension reduction and classification of hyperdimensional datasets. The method uses multivariate analysis to choose and apply a transform matrix that the transformed components are as orthogonal as possible. This nonlinear framework is derived by means of the theory of complete orthonormal systems. KMVSSA exhibits great flexibility by the combination of spectral and spatial features. We investigate the possibility of using KMVSSA for the classification of hyperspectral images and dimension reduction. The proposed framework is examined and compared in different merits with several hyperspectral images in different conditions (urban/agricultural area and size of the training set). Experimental results show that the proposed framework can meaningfully enhance the dimensionality reduction and also it greatly improves the overall as well as per class classification accuracies. We demonstrate a comprehensive comparison of some state of the art hyperspectral image classification methods.Index Terms-Airborne-satellite remote sensing, composite spectral-spatial kernels, dimension reduction, hyperspectral image classification, multivariate analysis, support vector machines (SVMs).
“…They follow virtual endmembers with the underlying assumption that all mixed pixels are encompassed by a minimum volume simplex. Moreover, several hyperspectral unmixing techniques have been proposed by identifying endmembers and their designated abundance fractions simultaneously like nonnegative matrix factorization (NMF) [13], [14], multilayer NMF (MLNMF) [15], minimum volume enclosing simplex (MVES) [16], and graph-regularized NMF (GNMF) method combined with sparseness constraint [17]. All the aforementioned strategies consider hyperspectral data cube as a cloud irregularly involving huge spectral vectors without any spatial arrangements.…”
Spectral mixture analysis (SMA) is an effective tool in recognition of unique spectral signatures of materials called endmembers and estimating their percentage of existence (abundance fractions). Most approaches designed in endmember extraction process are established by applying the spectral information of the dataset and, thus, tend to neglect the existing spatial correlation between adjacent pixels. Although several preprocessing modules have been developed by incorporating both spatial and spectral properties prior to spectral-based endmember extraction algorithms (EEs), they still encounter several challenges. Hence, in this paper, we propose an appropriate clustering and oversegmentation-based preprocessing (COPP) by greatly benefiting from the integration of spatial and spectral information. Moreover, a novel top-down oversegmentation (TDOS) algorithm is developed which can recognize small oversegments with high spatial correlation. Our scheme removes oversegments located at spatial border of cluster regions. Average spectral vectors of determined spatially homogenous oversegments are considered so that their spectral purity scores are calculated. COPP identifies spatially homogenous zones with the greatest spectral purity scores. Pixels of these regions are more likely to be adopted as endmembers by means of subsequent EEs. COPP can take advantage of degrading local spectral variability and noise power. The main contribution of this paper is the enhanced computational performance of EE as well as the precise reconstruction of the original hyperspectral scene besides its appropriate recognition of endmembers' spectral signatures. The effectiveness of our design and its validation are appraised with the state-of-the-art strategies on a synthetic and AVIRIS real hyperspectral datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.