Combinatorial Various Classification Proceedings for Hyper spectral Evidence Awareness

Author(s):

M.Divyabharathi , Christian college of Engineering And Technology, Dindigul , Tamilnadu; P.Pavithra, Christian college of Engineering And Technology, Dindigul , Tamilnadu; S.Sobiya, Christian college of Engineering And Technology, Dindigul , Tamilnadu; S. Nithya Priya, Christian college of Engineering And Technology, Dindigul , Tamilnadu

Keywords:

Genetic Algorithm, Hyperspectral Image, Support Vector Machine, Feature Extraction, Pixel Representation, Land Cover Classification

Abstract:

In land cover classification, Hyperspectral image investigation have been used in the field of remote sensing. Hyperspectral means hundreds of bands In this classification of hyperspectral we aim to produce a thematic map is more accurate by combinatorial classification methods. A hyperspectral data is a raw data which was undetermined. The feature representation is based on two learning algorithms (Support Vector Machine (SVM) and Artificial Neural Network (ANN)) were used to perform the combination function. In this work the main aim to produce the thematic map for survey analysis which is easy to the employee can work anywhere at any time. In this method the genetic algorithm(GA) is also used to produce more accuracy in thematic map. Here the dotnet painting tool is used for color representation. Our proposal was able to reduce the time complexity , no need for assigning weights and overcome the difficulties of the usual combination rules.

Other Details:

Manuscript Id	:	IJSTEV2I10047
Published in	:	Volume : 2, Issue : 10
Publication Date	:	01/05/2016
Page(s)	:	170-173

Download Article