Paper Title
Decision Level Fusion Of High Resolution Satellite Data For Urban Area Analysis

In the field of multi-source data fusion, fusion of multispectral and panchromatic remote sensing data in urban area has attracted more attention. Multi-source data has been remarkably increased for classification. This is because, the different sources may provide more information, and fusion of different information can produce a better understanding of the observed site. This paper addressed the use of a decision fusion methodology for the combination of multispectral and panchromatic data in urban area. The proposed method applied a support vector machine (svm)-based classifier fusion system for fusion of the multi-source data in the decision level. First, radiometric feature are extracted on multispectral data. Then, svm based rbf kernel classifiers are applied on each feature data, and on the panchromatic data. After producing multiple of classifiers, two comparative data fusion techniques are applied as a classifier fusion method to combine the results of svm classifiers form the data sets. Experimental results show that the proposed data fusion method improved the classification accuracy and kappa coefficient in comparison to the single data sets. The results revealed that the overall accuracies of svm classification on the multi specral and the panchromatic data separately are 60.3% and 59.1%, while our decision fusion methodology with majority voting technique receive the accuracy up to 88.6%, and dempster shafer receive the accuracy up to 94.7%.