Discrimination Features by Wavelet Transformation

Mustafa, Faten A.; Khalil, Turkan A.; Khalil, Almas A.

doi:10.33899/csmj.2009.163814

	Discrimination Features by Wavelet Transformation
AL-Rafidain Journal of Computer Sciences and Mathematics
Article 22, Volume 6, Issue 2, 2009, Pages 33-46 PDF (742.02 K)
Document Type: Research Paper
DOI: 10.33899/csmj.2009.163814
Authors
Faten A. Mustafa¹; Turkan A. Khalil²; Almas A. Khalil³
¹Remote Sensitivity Center
²College of Engineering University of Mosul
³College of Basic Education
Abstract
Recently, techniques and applications in space data processing it's and study characteristics in different areas and applications such as remote sensing study of land use, land cover and geological applications are increased. The research contained study and application of technical wavelet transformation to highlight information packages statement space for Mosul Dam and extracting ground obvious manifestations of this region by applying the classification on the wave beam generated. The results showed that the classification of images resulting from the wavelet transformation was accurate and of high amount (%96.715) compared to the results of the classification of non-original images using the improved wavelet transformation, with the accuracy of the classification (%92.652) and to obtain a clear picture parameters for the study of high quality. especially when discrimination features such as terrestrial lake, the body of dams and residential areas in addition to discrimination agricultural areas, the classification algorithm, which was used in this study, is an algorithm of Mahalanobis Distance.
Keywords
Discrimination Features; Wavelet transformation; Algorithm of Mahalanobis Distance

Statistics Article View: 273 PDF Download: 1,243