期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2017
卷号：17
期号：4
页码：224-237
出版社：International Journal of Computer Science and Network Security
摘要：Urbanization is the major strategy that leads to the loss of the various biodiversity and the homogenization of geological habitat. Thus, it is necessary to study and analyze the drastic changes occurred due to global urbanization periodically. The periodical assessment of urbanization gives rise to the development of various techniques and rules from several researchers. This paper is also a part of development over the urban-land cover analysis. It introduces a Budget in Random Forest Decision tree (RFDT) approach that preserves the statistical features and object boundaries and help in improving the classification accuracy. The system implements a spectral band segmentation method, which differentiates the remote sensory images as Land Cover (LC) and Land Use (LU). The proposed RF algorithm of decision trees attempts to provide an improved efficiency over the other existing methods that is obtained from the experimental verification of the earlier algorithms with the proposed. The comparison report shows the performance of the algorithm from 2007 to 2013 respectively.
关键词：Decision Trees; Geographic Information Systems; Land Cover; Land Use; Random Forest Decision Tree.
