An improved maximum simplex volume algorithm to unmixing hyperspectral data

Haicheng Qu; Bormin Huang; Junping Zhang; Ye Zhang

doi:10.1117/12.2034759

23 October 2013 An improved maximum simplex volume algorithm to unmixing hyperspectral data

Haicheng Qu, Bormin Huang, Junping Zhang, Ye Zhang

Author Affiliations +

Proceedings Volume 8895, High-Performance Computing in Remote Sensing III; 889507 (2013) https://doi.org/10.1117/12.2034759
Event: SPIE Remote Sensing, 2013, Dresden, Germany

Abstract

The maximum simplex volume algorithm (MSVA) is an automatic endmember extraction method based on geometrical properties of simplex in high-dimensional feature space. By utilizing the relation of volume between a simplex and its corresponding parallelohedron in the high-dimensional space, the algorithm extracts endmembers directly from the initial hyperspectral image in a sequential manner without dimensionality reduction. It is thus considered to have overcome a major drawback of N-FINDR algorithm, which requires the data dimension reduced to one less than the number of the endmembers before endmembers extraction. But the MSVA suffers from excessive computation caused by massive determinant operation in practical application. An improved fast implementation method based on partitioned determinant operation is proposed in this paper to reduce computational complexity. Experimental results demonstrate that the proposed fast algorithm can greatly reduce computational complexity, while simultaneously their computational accuracyremains as good as its original algorithm.

Citation Download Citation

Haicheng Qu, Bormin Huang, Junping Zhang, and Ye Zhang "An improved maximum simplex volume algorithm to unmixing hyperspectral data", Proc. SPIE 8895, High-Performance Computing in Remote Sensing III, 889507 (23 October 2013); https://doi.org/10.1117/12.2034759

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