Land Cover Characterization Using Evolutionary and Adaptive Computing

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" Land Cover Characterization Using Evolutionary and Adaptive Computing "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	149691
Doc. No	:	ET21483
Main Entry	:	William M. Stiteler IV
Title Proper	:	Land Cover Characterization Using Evolutionary and Adaptive Computing
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	The classification of land cover using remotely-sensed imagery is a complicatedproblem, particularly when it involves several spectrally similar land cover types. Varioustypes of information potentially useful to the classification problem are not used by mosttraditional methods of image classification. For example, a maximum likelihoodclassification looks at only the spectral characteristics of a single pixel, across several bands,ignoring a great deal of spatial information (ERDAS, 1999). Traditional, purely spectralmethods assume that there are hard, physical manifestations of differences between semanticland cover classes, and that a given pixel contains one and only one land cover class.Because of these assumptions, purely spectral methods work well in areas of homogeneousland cover, but often fail in heterogeneous areas. Image pixels that occur withinhomogeneous areas tend to be clear examples of a given class, and generally exhibit lesswithin-pixel variation than pixels that exist in more heterogeneous areas.tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL44653.pdf
Title and statement of responsibility and	:	Land Cover Characterization Using Evolutionary and Adaptive Computing [Thesis]