&lt;title&gt;Spectral unmixing of remotely sensed imagery using maximum entropy&lt;/title&gt;

Chettri, Samir R.; Netanyahu, Nathan S.

doi:10.1117/12.267839

Cited by 12 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4) Sparse Selection: According to (11), if d (k) < T d , the corresponding spectrum will be selected for the kth class; otherwise, there is no endmember selected for the kth class.…”

Section: Algorithm 1 the Sses Algorithmmentioning

confidence: 99%

“…In the past decades, a large number of algorithms has been proposed for these two steps, such as N-FINDR algorithm [4], vertex component analysis (VCA) [5], simplex growing algorithm (SGA) [6], automated morphological endmember extraction (AMEE) algorithm [7], [8], and spatial purity-based endmember extraction (SPEE) algorithm [9], [10] for EE; gradient descent maximum entropy (GDME) algorithm [11], fully constrained least square (FCLS) algorithm [12], and multichannel hopfield neural network (MHNN) [2], [13] for AE. In addition, unsupervised unmixing algorithms, which usually perform EE and AE simultaneously, have also widely been researched, such as nonnegative matrix factorization based algorithms [14]- [17], convex optimization based algorithms [18]- [20], to name a few.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Equivalent-Sparse Unmixing Through Spatial and Spectral Constrained Endmember Selection From an Image-Derived Spectral Library

Mei

2015

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

View full text Add to dashboard Cite

Spectral variation, which is inevitably present in hyperspectral data due to nonuniformity and inconsistency of illumination, may result in considerable difficulty in spectral unmixing. In this paper, a field endmember library is constructed to accommodate spectral variation by representing each endmember class by a batch of image-derived spectra. In order to perform unmixing by such a field endmember library, a novel spatial and spectral endmember selection (SSES) algorithm is designed to search for a spatial and spectral constrained endmember subset per pixel for abundance estimation (AE). The net effect is to achieve sparse unmixing equivalently, considering the fact that only a few endmembers in the large library have nonzero abundances. Thus, the resulting algorithm is called spatial and spectral constrained sparse unmixing (SSCSU). Experimental results using both synthetic and real hyperspectral images demonstrate that the proposed SSCSU algorithm not only improves the performance of traditional AE algorithms by considering spectral variation, but also outperforms the existing sparse unmixing approaches.Index Terms-Hyperspectral image, in-field spectral variation, mixed pixel, sparse unmixing, spectral unmixing.

show abstract

“…4) Sparse Selection: According to (11), if d (k) < T d , the corresponding spectrum will be selected for the kth class; otherwise, there is no endmember selected for the kth class.…”

Section: Algorithm 1 the Sses Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Equivalent-Sparse Unmixing Through Spatial and Spectral Constrained Endmember Selection From an Image-Derived Spectral Library

Mei

2015

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

View full text Add to dashboard Cite

show abstract

“…To take into account the local variation of end-member spectra, Chettri and others (1997) suggest using an average spectrum of several pixels derived from the extremes of the polygon, instead of a single spectrum value.We calculate a mean average of all potential snow spectra, taking into account the different snow reflectances which result from variations in elevation and exposure of snow surfaces in alpine terrain (Painter and others, 1998) and from minimal end-member impurities. As seen in Figure 2, the snow end-members show some variability of their spectra, and the greatest differences occur in the near-infrared (channel 2), which is sensitive to moderate amounts of impurities in the snow because absorbing particulates affect snow reflectance out to 0.9 μ m (Grenfell and others, 1981).…”

Section: Algorithmmentioning

confidence: 99%

Operational sub-pixel snow mapping over the Alps with NOAA AVHRR data

et al. 2004

View full text Add to dashboard Cite

This study is part of research activities concentrating on the real-time application of the U.S. National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) sensor for snow-cover analysis of the European Alps. For mapping snow cover in heterogeneous terrain, we implement the widely used linear spectral mixture algorithm to estimate snow cover at sub-pixel scale. Principal component analysis, including the reflective part of AVHRR channel 3, is used to estimate fractions of “snow” and “not snow” within a pixel, using linear mixture modeling. The combination of these features leads to a fast, simple solution for operational and near-real-time processing. The presented algorithm is applied on the European Alps on 17 January 2003 and successfully maps snow at sub-pixel scale. The detailed snow-cover information makes it easy to recognize the complex topography of the Alps, more so than with either a classic binary map or a Moderate Resolution Imaging Spectroradiometer (MODIS) snow product. The sub-pixel algorithm reasonably identifies snow-cover fractions in regions and at altitudes where neither the classic binary map nor the MODIS algorithm detects any snow. Differences concerning the snow distribution are found in forested areas as well as in the lowest-elevation zones. The algorithm substantially improves snow mapping over complex topography for operational and near-realtime applications.

show abstract

“…We term this algorithm as unsupervised maximum entropy (uMaxEnt). Although the maximum entropy (MaxEnt) principle has been applied to hyperspectral data unmixing previously [1,3], these algorithms either did not present a practical solution to endmember detection or did not provide a set of comprehensive experimental results to valid the effectiveness of the algorithm. In addition, these papers lack comparative analysis with other available methods using popular test data.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Decomposition of Mixed Pixels Using the Maximum Entropy Principle

Miao

Szu

2006

18th International Conference on Pattern Recognition (ICPR'06)

View full text Add to dashboard Cite

Due to the wide existence of mixed pixels, the derivation of constituent components (endmembers) and their proportions (abundances) at subpixel scales has become an important research topic. In this paper, we propose a novel unsupervised decomposition method based on the classical maximum entropy principle, termed uMaxEnt. The algorithm integrates a global least square error-based endmember detection and a per-pixel maximum entropy learning to find the most possible proportions. We apply the proposed method to the subject of spectral unmixing. The experimental results obtained from both simulated and real hyperspectral data demonstrate the effectiveness of the uMaxEnt method.

show abstract

<title>Spectral unmixing of remotely sensed imagery using maximum entropy</title>

Cited by 12 publications

References 0 publications

Equivalent-Sparse Unmixing Through Spatial and Spectral Constrained Endmember Selection From an Image-Derived Spectral Library

Equivalent-Sparse Unmixing Through Spatial and Spectral Constrained Endmember Selection From an Image-Derived Spectral Library

Operational sub-pixel snow mapping over the Alps with NOAA AVHRR data

Unsupervised Decomposition of Mixed Pixels Using the Maximum Entropy Principle

Contact Info

Product

Resources

About