Feature Extraction Based Hybrid Classifier for Classifying Remote Sensing Images
M. Praneesh1, D. Napoleon2
1M. Praneesh, Department of Computer Science, Bharthiar University, Sri Ramakrishna College of Arts and Science, Coimbatore, India.
2Dr. D. Napoleon, Department of Computer Science, Bharthiar University, Coimbatore, India.

Manuscript received on 09 April 2019 | Revised Manuscript received on 14 May 2019 | Manuscript published on 30 May 2019 | PP: 1636-1639 | Volume-8 Issue-1, May 2019 | Retrieval Number: A9286058119/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Classification Techniques are encompassed on enormous databases to abridge models depicting different data classes. Advantageously, such kind of analysis can render a deep-seated perceptivity for appropriate understanding of different large-scale databases. Studies related to acquaintance and developments of knowledge are also very proficient and are one of the first and foremost utility in the remote sensing field with satellite imagery datasets. The decision making process in any remote sensing research is predominantly bet on the effectiveness of the classification process. In order to identify six land type classes, efficient classification techniques were developed and embraced to a landsat satellite database inculcated with Irvine machine learning repository at university of California. The ultimate intention of this paper is to guesstimate and take account of the proficient performance of proposed algorithm (Hybrid GASVM) in the analysis of the classified lands from this large set of satellite imaginary and also compared proposed algorithm with traditional classifier algorithm like Multilayer perception back propagation neural network, support vector machine and K-Nearest neighbor. In accordance to measure the classification accuracy, Average producer accuracy, Average user accuracy, kappa statistic, various performance measures were applied.
Index Terms: Image Classification, Machine Learning Algorithm, Confusion Matrix, Unsupervised Learning

Scope of the Article: Classification