Accuracy assessment of rough set based SVM technique for spatial image classification. (2018)

• Record Type: Journal Article
• Authors: Vasundhara, D.N.; Seetha, M.
• Is Part Of: International journal of knowledge and learning. Volume 12:Number 3(2018)
• Page Start: 269
• Page End: 285
• Publication Date: 2018
• Subjects: feature extraction -- classification -- rough sets -- ANN -- artificial neural network -- support vector machines; Knowledge and learning -- Periodicals; Knowledge management -- Periodicals; 306.4205
• Languages: English
• ISSNs: 1741-1009

Abstract

There exist many traditional spatial image classification techniques which are developed over past years and exists in literature. Today, expert systems along with machine learning methods are getting universality in this area due to effective classification. This paper presents Rough set based support vector machine (SVM) classification (RS-SVM) method. In this technique, Rough set (RS) is used as a feature selection mathematical tool which eliminates the redundant features. Further, this reduced dimensionality dataset is given to SVM classifier. This process improves the classification accuracy and performance. We have performed experiments using standard geospatial images for above-proposed method for classification.