Impact factor analysis of mixture spectra unmixing based on independent component analysis

Lei Chen; Shengbo Chen; Xulin Guo; Chao Zhou

doi:10.1117/1.JRS.10.015012

15 February 2016 Impact factor analysis of mixture spectra unmixing based on independent component analysis

Lei Chen, Shengbo Chen, Xulin Guo, Chao Zhou

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 10, Issue 1, 015012 (February 2016). https://doi.org/10.1117/1.JRS.10.015012

Abstract

Based on spectral independence of different materials, independent component analysis (ICA), a blind source separation technique, can be applied to separate mixed hyperspectral signals. For the purpose of detecting objects on the sea and improving the precision of target recognition, an original ICA method is applied by analyzing the influence exerted by spectral features of different materials and mixture materials on spectral unmixing results. Due to the complexity of targets on the sea, several measured spectra of different materials have been mixed with water spectra to simulate mixed spectra for mixture spectra decomposition. Synthetic mixed spectra are generated by linear combinations of different materials and water spectra to obtain separated results. We then compared the separated results with the measured spectra of each endmember by coefficient of determination. We conclude that these factors that will change the original spectral characteristics of Gaussian distribution have significant influence on the separated results and selecting a proper initial matrix, and processing spectral data with lower noise can help improve the ICA method for more accurate separated results from hyperspectral data.

Citation Download Citation

Lei Chen, Shengbo Chen, Xulin Guo, and Chao Zhou "Impact factor analysis of mixture spectra unmixing based on independent component analysis," Journal of Applied Remote Sensing 10(1), 015012 (15 February 2016). https://doi.org/10.1117/1.JRS.10.015012

Published: 15 February 2016

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

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.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available