Hyperspectral Image Analysis Using Genetic
Programming
B. Ross, A. Gualtieri, F. Fueten, and P. Budkewitsch. GECCO 2002: Proceedings of the Genetic and
Evolutionary Computation Conference, page 1196--1203. New York, Morgan Kaufmann Publishers, (9-13 July 2002)
Abstract
Genetic programming is used to evolve mineral
identification functions for hyperspectral images. The
input image set comprises 168 images from di#erent
wavelengths ranging from 428 nm (visible blue) to 2507
nm (invisible shortwave in the infrared), taken over
Cuprite, Nevada, with the AVIRIS hyperspectral sensor.
A composite mineral image indicating the overall
reflectance percentage of three minerals (alunite,
kaolnite, buddingtonite) is used as a reference or
"solution " image. The training set is manually
selected from this composite image. The task of the GP
system is to evolve mineral identifiers, where each
identifier is trained to identify one of the three
mineral specimens. A number of di#erent GP experiments
were undertaken, which parameterized features such as
thresholded mineral reflectance intensity and target GP
language. The results are promising, especially for
minerals with higher reflectance thresholds (more
intense concentrations).
GECCO 2002: Proceedings of the Genetic and
Evolutionary Computation Conference
year
2002
month
9-13 July
pages
1196--1203
publisher
Morgan Kaufmann Publishers
publisher_address
San Francisco, CA 94104, USA
isbn
1-55860-878-8
notes
GECCO-2002. A joint meeting of the eleventh
International Conference on Genetic Algorithms
(ICGA-2002) and the seventh Annual Genetic Programming
Conference (GP-2002) See also
ross2:2002:geccoTR
%0 Conference Paper
%1 ross2:2002:gecco
%A Ross, Brian J.
%A Gualtieri, Anthony G.
%A Fueten, Frank
%A Budkewitsch, Paul
%B GECCO 2002: Proceedings of the Genetic and
Evolutionary Computation Conference
%C New York
%D 2002
%E Langdon, W. B.
%E Cantú-Paz, E.
%E Mathias, K.
%E Roy, R.
%E Davis, D.
%E Poli, R.
%E Balakrishnan, K.
%E Honavar, V.
%E Rudolph, G.
%E Wegener, J.
%E Bull, L.
%E Potter, M. A.
%E Schultz, A. C.
%E Miller, J. F.
%E Burke, E.
%E Jonoska, N.
%I Morgan Kaufmann Publishers
%K algorithms, applications, classification genetic hyperspectral imaging, mineral programming, real world
%P 1196--1203
%T Hyperspectral Image Analysis Using Genetic
Programming
%U http://citeseer.ist.psu.edu/503556.html
%X Genetic programming is used to evolve mineral
identification functions for hyperspectral images. The
input image set comprises 168 images from di#erent
wavelengths ranging from 428 nm (visible blue) to 2507
nm (invisible shortwave in the infrared), taken over
Cuprite, Nevada, with the AVIRIS hyperspectral sensor.
A composite mineral image indicating the overall
reflectance percentage of three minerals (alunite,
kaolnite, buddingtonite) is used as a reference or
"solution " image. The training set is manually
selected from this composite image. The task of the GP
system is to evolve mineral identifiers, where each
identifier is trained to identify one of the three
mineral specimens. A number of di#erent GP experiments
were undertaken, which parameterized features such as
thresholded mineral reflectance intensity and target GP
language. The results are promising, especially for
minerals with higher reflectance thresholds (more
intense concentrations).
%@ 1-55860-878-8
@inproceedings{ross2:2002:gecco,
abstract = {Genetic programming is used to evolve mineral
identification functions for hyperspectral images. The
input image set comprises 168 images from di#erent
wavelengths ranging from 428 nm (visible blue) to 2507
nm (invisible shortwave in the infrared), taken over
Cuprite, Nevada, with the AVIRIS hyperspectral sensor.
A composite mineral image indicating the overall
reflectance percentage of three minerals (alunite,
kaolnite, buddingtonite) is used as a reference or
{"}solution {"} image. The training set is manually
selected from this composite image. The task of the GP
system is to evolve mineral identifiers, where each
identifier is trained to identify one of the three
mineral specimens. A number of di#erent GP experiments
were undertaken, which parameterized features such as
thresholded mineral reflectance intensity and target GP
language. The results are promising, especially for
minerals with higher reflectance thresholds (more
intense concentrations).},
added-at = {2008-06-19T17:35:00.000+0200},
address = {New York},
author = {Ross, Brian J. and Gualtieri, Anthony G. and Fueten, Frank and Budkewitsch, Paul},
biburl = {https://www.bibsonomy.org/bibtex/260cb6b574d02444f6488090e730d4c1f/brazovayeye},
booktitle = {GECCO 2002: Proceedings of the Genetic and
Evolutionary Computation Conference},
editor = {Langdon, W. B. and Cant{\'u}-Paz, E. and Mathias, K. and Roy, R. and Davis, D. and Poli, R. and Balakrishnan, K. and Honavar, V. and Rudolph, G. and Wegener, J. and Bull, L. and Potter, M. A. and Schultz, A. C. and Miller, J. F. and Burke, E. and Jonoska, N.},
interhash = {a2d9c57bf6670355422dbdf3f5e50126},
intrahash = {60cb6b574d02444f6488090e730d4c1f},
isbn = {1-55860-878-8},
keywords = {algorithms, applications, classification genetic hyperspectral imaging, mineral programming, real world},
month = {9-13 July},
notes = {GECCO-2002. A joint meeting of the eleventh
International Conference on Genetic Algorithms
(ICGA-2002) and the seventh Annual Genetic Programming
Conference (GP-2002) See also
\cite{ross2:2002:geccoTR}},
pages = {1196--1203},
publisher = {Morgan Kaufmann Publishers},
publisher_address = {San Francisco, CA 94104, USA},
timestamp = {2008-06-19T17:50:41.000+0200},
title = {Hyperspectral Image Analysis Using Genetic
Programming},
url = {http://citeseer.ist.psu.edu/503556.html},
year = 2002
}