期刊名称:International Journal of Advanced Computer Science and Applications(IJACSA)
印刷版ISSN:2158-107X
电子版ISSN:2156-5570
出版年度:2020
卷号:11
期号:10
DOI:10.14569/IJACSA.2020.0111087
出版社:Science and Information Society (SAI)
摘要:National and regional infrastructure planning is founded on the use of many factors, of which population size can be argued to be the most fundamental. Population size is typically acquired through a census. However, manual census collection is an expensive and resource intensive process; especially in regions that are poorly connected. Computer-aided population estimation, when done accurately, therefore offers significant benefit. This paper presents a comprehensive framework for estimating the population size of a region of interest by applying classification techniques to terrain data. Central to the proposed framework is a novel histogram representation technique designed to support the generation of appropriate and effective classifiers central to the operation of the framework. The presented work uses the Karak region, in Jordan, as a case study for population size estimation. The proposed framework and the representation technique have been evaluated using a variety of classification mechanisms and parameter settings. The reported evaluation of the proposed representation technique demonstrates that good results can be obtained with regard to estimate the population size.
关键词:Histogram representation; Geographic Information System (GIS); population estimation; 3D surface; satellite images; data mining; classification technique; Karak region