期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2014
卷号：66
期号：3
出版社：Journal of Theoretical and Applied
摘要：Due to the dramatic change of air temperature, the ice layer expands significantly in the lake region. To monitor and to predict the ice conditions occurred in Ice Lake over seasonal basis, the pixel based classification method is evaluated and experimented. Research in classification of lake ice concurrently takes the history of ice information and structure of the ice cover. In this paper, based on the fundamental information, a methodology is proposed to classify ice from other surfaces based on pixel intensities. By applying Color based segmentation using K-Mean Cluster the intensity of different classes are extracted according to the respective ranges. The pixel intensity ranges are taken as a feature value for the KNN Classification. With that pixel intensity range three classes are identified as ice, water and sand. Experimental results show that the growth rate of the ice is estimated with the classified ice region in the lake area .The accuracy rate of ice classification from other classes in an image is high using KNN. This examination explores and proves the expansion of the ice cover according to the seasonal duration using image analysis methodology.
