Title: Accuracy assessment of rough set based SVM technique for spatial image classification

Authors: D.N. Vasundhara; M. Seetha

Addresses: Department of C.S.E., VNR Vignana Jyothi Institute of Engineering and Technology Hyderabad, 500090, India ' Department of C.S.E. G Narayanamma Institute of Technology and Science, Hyderabad, 500104, India

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.

Keywords: feature extraction; classification; rough sets; ANN; artificial neural network; support vector machines.

DOI: 10.1504/IJKL.2018.092318

International Journal of Knowledge and Learning, 2018 Vol.12 No.3, pp.269 - 285

Received: 18 Mar 2017
Accepted: 16 Jan 2018
Published online: 14 Jun 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article