Thesis (M.Sc.) - Cairo University - Faculty Of Computers and information - Department Of Information Technology

Most of the change detection techniques may be categorized under one of two basic categories : comparative analysis techniques , iepost classification change detection ; and simultaneous analysis techniques , ieimage to image change detection or differencing techniquesA simple and intuitively attractive technique to the detection of changes in imagery is therefore to independently classify each image , and then compare their class labelsThis technique is known as post classification comparisonIn this thesis we introduce a technique to improve the change detection capability for remote sensing images using fuzzy logicIn our technique we attempt to overcome the limitations of the post classification comparison technique by involving an independent fuzzy classification of digital SPOT satellite images from different dates then posing logical queries over those classified imagesUsing fuzzy logic for post classification comparison provides partial class assignment in regions where there is a gradual transition from one class to anotherWe undertake the analysis of the changes to find the criteria to improve change detection capabilityAlso our technique not only detects changes and their nature but also can detect the orientation of such changesThe proposed technique is implemented on SPOT subscene covering El - Manzala lake in northern Egypt detecting the changes in the period between 1990 - 2000

Issued also as CD