KEYWORDS: Expectation maximization algorithms, Data modeling, Algorithm development, Performance modeling, Minerals, Statistical analysis, Monte Carlo methods, Computer simulations, Statistical modeling, Signal to noise ratio
The independent component analysis has been commonly employed in hyperspectral unmixing. However, the success of this method is highly dependent on the independency of its sources assumption. Dependent component analysis (DECA) algorithm, which utilizes a Dirichlet mixture model, was developed to provide more adequate spectral unmixing of dependent sources. Estimation of the unknown model parameters using the expectation maximization algorithm in DECA resulted in some insufficiencies. DECAGibbs algorithm is introduced to improve unmixing accuracy by applying the Gibbs sampling method to the parameter estimation process of DECA, which is conducted in different manners of modeling the observations. Functionality of the DECAGibbs algorithm is examined through the artificial datasets and an AVIRIS image of Cuprite, Nevada, indicating better decomposition of mixed observations. Finally, the best performing model was employed in mineralogical mapping of the Lahroud region, northwest Iran, by a Hyperion image. The results represent the high reliability of the proposed method according to the geological studies of the area. Since the practical application of the mixture models relies upon the efficient estimation of their involved parameters, the performance of the DECA algorithm is improved by employing the Bayesian parameter estimation approaches in this research.
Mapping of alterations in a geological terrain can be considered as a classification task in the remote sensing data
processing. Training dataset is an important part of a classification process. Collecting of precise training data is
generally expensive and time consuming. In this study, the alteration map resulted by Hyperion is used as training data
for classification of the ASTER scene in Erongo, Namibia. This extends results to a much broader in comparison to
Hyperion scene. Ten alterations detected by the matched filtering unmixing method on the Hyperion dataset are therefore
training classes of the classification. The separability of the classes was computed to evaluate the ability of ASTER data
to spectrally discriminate between these classes. The outcome of this computation is satisfactory for the high-probability
training dataset. In order to improve the accuracy of upcoming processes, classes with high similarity (low separability)
were combined. The classification of ASTER scene is then performed with the use of both individual and combined
classifiers. An accuracy analysis was performed to compare the accuracy of each classifier. The Mahalanobis distance
method has the best performance among all classifiers regarding to its highest overall accuracy.
The study aims to detect and map alteration halos in Erongo Namibia using Hyperion dataset. Detail surveys and
investigations are possible considering hyperspectral sensors capability to render plenty of spectral information
from observing surface of earth. In the term of mineral detection there are particular challenges. Main problem is
due to very small size of mineral grain comparing to even those data possessing finest ground resolution.
One of the methods has been invented for objects smaller than ground resolution of dataset is linear mixture
model (LMM). This allows us to estimate abundances of targets in each pixel of scene after determination of
endmembers. Regarding to the challenge about mineral detection task, as a matter of fact, finding pure pixels of
the data in mineralogical scale is often impossible. Therefore we are able to determine the purest pixels which are
a mixture of several minerals themselves and spectral profile of them consists of absorption features of those
detectable minerals.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.