By Topic

An application of genetic algorithms on band selection for hyperspectral image classification

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Ji-Ping Ma ; Sch. of Remote Sensing Inf. Eng., Wuhan Univ., China ; Zhao-Bao Zheng ; Qing-Xi Tong ; Lan-Fen Zheng

Selection of optimal bands is an effective means to mitigate the curse of dimensionality for remotely sensed data, which has assumed growing importance with the availability of hyperspectral remote sensing data. For the merits of genetic algorithms in solving problems of combinatorial optimization, by using special coding mechanism and fitness function, this paper develops an approach to fast search of optimal bands for hyperspectral image data classification. Experiments demonstrate that this method can automatically obtain the best combination with much faster speed as compared with a conventional overall search method, and that by comparison, three-band combination search process exhibits much more efficiency in comparison with conventional methods than two-band combination does, which indicates that with more bands in each combination, more obvious the efficiency of this method is. By using the selected optimal bands for maximum likelihood classification of the three vegetation types in the experiment area, satisfied result is achieved. Because of no knowledge of interesting objects is required to use this method, it has the merit of being suitable for unknown areas.

Published in:

Machine Learning and Cybernetics, 2003 International Conference on  (Volume:5 )

Date of Conference:

2-5 Nov. 2003