GEOSCAN Search Results: Fastlink


TitleHyperspectral image analysis using genetic programming
AuthorRoss, B J; Gualtieri, A G; Fueten, F; Budkewitsch, P
SourceApplied Soft Computing vol. 5, issue 2, 2005 p. 147-156, Open Access logo Open Access
LinksAbstract - Résumé
Alt SeriesEarth Sciences Sector, Contribution Series 2004300
PublisherElsevier BV
Mediapaper; on-line; digital
File formatpdf
AreaCuprite, Nevada; United States of America
Lat/Long WENS-118.0000 -117.0000 38.0000 37.0000
Subjectsgeochemistry; geophysics; spectral analyses; reflectance; satellite imagery; infrared spectral analyses; LANDSAT imagery
Illustrationsdigital images; spectra; tables; satellite images; formulae; photomicrographs
ProgramNSERC Natural Sciences and Engineering Research Council of Canada
AbstractGenetic programming is used to evolve mineral identification functions for hyperspectral images. The input image set comprises 168 images from different 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, and results are cross-validated with the remaining image data not used for training. 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 different 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, which indicate more intense concentrations.

Date modified: