Decision Tree Approach for Classification of Satellite Imagery
Havyas V B1, Choodarathnakara A L2, Thribhuvan R3, Chethan K S4
1Prof. Havyas V B, Department of Electronics & Communication Engineering, Government Engineering College, Kushalnagar, Kodagu, INDIA.
2Prof. A L Choodarathnakara, Department of Electronics & Communication Engineering, Government Engineering College, Kushalnagar, Kodagu, IND IA.
3Prof. Thribhuvan R, Department of Electronics & Communication Engineering, Coorg Institute of Technology, Ponnampet, Kodagu, INDIA.
4Prof. Chethan K S, Department of Electronics & Communication Engineering, PESIT, Bangalore, INDIA.
Manuscript received on April 20, 2015. | Revised Manuscript received on April 28, 2015. | Manuscript published on March 05, 2015. | PP: 101-104 | Volume-5, Issue-2, May 2015. | Retrieval Number: B2614055215/2015©BEIESP
Open Access | Ethics and Policies | Cite
©The Authors. Published By: 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: Various practical systems capable of extracting descriptive decision making knowledge from data have been developed and evaluated. Techniques that represent knowledge about classification tasks in the form of decision trees are focused on. A sample of techniques is sketched, ranging from basic methods of constructing decision trees to ways of using them non-categorically. Some characteristics that suggest whether a particular classification task is likely to he amenable or otherwise to tree-based methods are discussed. Many urban land cover types show spectral similarity in remote sensing data. Further, the finer the spatial resolution of the data, the larger is the number of detectable subclasses within classes. This high within-class spectral variance of some classes results in multimodal distribution of spectra and may decrease their spectral separability. Hence, the existing traditional hard classification techniques which are parametric type do not perform well on high resolution data in the complex environment of the urban area as they expect datasets to be distributed normally. The aim of this paper is to investigate a non-parametric classifier as an alternative approach to classify an image data of a semi urban area.
Keywords: Remote Sensing, Image Classification, Parametric Classifier, Non-parametric and Decision Tree Classifier.