Unsupervised Decomposition of Mixed Pixels Using the Maximum Entropy Principle

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 pr...

Full description

Saved in:
Bibliographic Details
Published in18th International Conference on Pattern Recognition (ICPR'06) Vol. 1; pp. 1067 - 1070
Main Authors Lidan Miao, Hairong Qi, Szu, H.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2006
Subjects
Area measurement
Entropy
Hyperspectral imaging
Hyperspectral sensors
Image analysis
Least squares methods
Matrix decomposition
Pixel
Remote sensing
Spatial resolution
Online AccessGet full text
ISBN0769525210
9780769525211
ISSN1051-4651
DOI10.1109/ICPR.2006.1142

Cover

More Information
Summary: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 hyper-spectral data demonstrate the effectiveness of the uMaxEnt method
ISBN:0769525210
9780769525211
ISSN:1051-4651
DOI:10.1109/ICPR.2006.1142