By Topic

Design and Implementation of Classification System for Satellite Images based on Soft Computing Techniques

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Kaghed, N.H. ; Dept. of Comput. Sci.,, Babylon Univ ; Abbas, T.A. ; Hussein Ali, S.

This paper presents a method to design programming system using hybrid techniques represented by soft computing to classify objects from the air photos and satellite images depending on their features with minimum acceptable error. These images usually consist of seven layers, while the work in this research focuses on dealing with three bands (red, green and blue). This paper concerns with classifying five kinds of objects (urban area, forests, roads, rivers, football-stadiums). Accordingly, the database which describes that objects depending on their attributes were built. Then, the Evolution algorithm of type breeder genetic algorithm to procedure genetic clustering process to segment image which provides a number of clusters found in that image data set were used. To avoid the overlapping between clusters with other, one of the clustering validity measures called "Davies-Bouldin index" as fitness function of that algorithm was used. Moreover, four methods of the recombination, which are:(discrete recombination (DR), extended line recombination (ELR), extended intermediate recombination (EIR), fuzzy recombination(FR)) were discussed. Then, two types of features for each cluster which are visual features including(pattern, shape, texture, shadow, associative), and statistical features represented by spectrum features that include (intensity, hue, daturation ) were extracted. After that, feed forward neural network from type error back propagation neural network to determine the class under which each feature vector belongs to was used. At the last stage, IF-Then rule to form several rules that govern each class attributes were used

Published in:

Information and Communication Technologies, 2006. ICTTA '06. 2nd  (Volume:1 )

Date of Conference:

0-0 0