Publication Type : Conference Paper
Source : Proceedings of the 2014 International Conference on Interdisciplinary Advances in Applied Computing, Association for Computing Machinery, New York, NY, USA (2014)
Url : https://doi.org/10.1145/2660859.2660920
ISBN : 9781450329088
Keywords : basis function, classification, codebook, Encoding, Feature extraction, pooling, Sparsity
Campus : Bengaluru
School : School of Engineering
Department : Electronics and Communication
Year : 2014
Abstract : A new method of classifying satellite images into different categories such as forest, desert, river etc., with the help of Support Vector Machine (SVM) and unsupervised learning method. In the past few decades satellite imagery has been used successfully for weather, geographical and geological applications. With the advance of technology, more sophisticated sensors provide higher resolutions, and with faster computer systems, the use of satellite imagery has opened the fields of exploration and application. In this paper we are going to use the unsupervised learning and Support Vector Machine in combination with Fisher's Linear Discriminate Analysis approach to classify the satellite images into the predefined categories. We use the algorithm to detect automatically the different objects like large shopping malls, facility areas, vegetation, ocean bodies etc.
Cite this Research Publication : Giriraja C. V., Haswanth, A., Srinivasa, C., JayaRam, T. K., and Krishnaiah, P., “Satellite Image Classification Using Unsupervised Learning and SIFT”, in Proceedings of the 2014 International Conference on Interdisciplinary Advances in Applied Computing, New York, NY, USA, 2014.