Hyperspectral image band selection based on genetic algorithm

Jiping Ma; Zhaobao Zheng; Qingxi Tong; Lanfen Zheng; Bin Zhang

doi:10.1117/12.441403

25 September 2001 Hyperspectral image band selection based on genetic algorithm

Jiping Ma, Zhaobao Zheng, Qingxi Tong, Lanfen Zheng, Bin Zhang

Proceedings Volume 4548, Multispectral and Hyperspectral Image Acquisition and Processing; (2001) https://doi.org/10.1117/12.441403
Event: Multispectral Image Processing and Pattern Recognition, 2001, Wuhan, China

Abstract

Optimum band selection for visual interpretation and classification is an interesting task in conventional remote sensing, and, as an effective means to mitigate the curse of dimensionality, which has assumed growing importance with the availability of hyperspectral remote sensing data. In determining three-channel combination for a informative display in an image-cube and determining feature combination for fast classification, band selection is regarded indispensable in hyperspectral remote sensing. When applied to data acquired from a hyperspectral sensor, which is usually with a set of hundreds of band, however, conventional band selection procedure, of any criterion, becomes not viable with respect to the particularly time consuming. To cope with this pitfall, a method based upon genetic algorithm is proposed in this paper. An experiment, with a 121 band data set, demonstrate the efficiency. For simplification, the algorithm is designed to choose a combination which produces the most informative visual result when used as the top color preference in an image- cube. With little modification in criterion, the algorithm can be used to select features for classification purpose. The corresponding result is also presented in this paper.

Citation Download Citation

Jiping Ma, Zhaobao Zheng, Qingxi Tong, Lanfen Zheng, and Bin Zhang "Hyperspectral image band selection based on genetic algorithm", Proc. SPIE 4548, Multispectral and Hyperspectral Image Acquisition and Processing, (25 September 2001); https://doi.org/10.1117/12.441403

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