[Objective] This study aimed to improve classification accuracy of RS images using rough set theory in the growth of crops. [Method] Technique methods of data mining and knowledge discovery have been used. The develop...[Objective] This study aimed to improve classification accuracy of RS images using rough set theory in the growth of crops. [Method] Technique methods of data mining and knowledge discovery have been used. The development status of spatial data mining and knowledge discovery （SDMKD） is presented and data mining techniques in remote sensing were deeply analyzed. Then, SDMKD of TM image are researched using method of rough set, mainly including four methods （rough set, apriori algorithms, inductive learning, clustering）. [Result] The proposed method raises efficiency of land use and land reclaim. Based on the SDMKD, the characteristics of TM showed that the information after using rough set is more intensive than that of none. Especially, much better results can be gained while kinds of corps are less than five. [Conclusion] This study laid significant basis for further research on data mining in the growth of crops.展开更多
基金Supported by the by Research Fund for the Doctoral Program of Higher Education of China(20096121120001)Science Research Program of Educational Commission of Shaanxi Province of China(12JK0781)~~
文摘[Objective] This study aimed to improve classification accuracy of RS images using rough set theory in the growth of crops. [Method] Technique methods of data mining and knowledge discovery have been used. The development status of spatial data mining and knowledge discovery （SDMKD） is presented and data mining techniques in remote sensing were deeply analyzed. Then, SDMKD of TM image are researched using method of rough set, mainly including four methods （rough set, apriori algorithms, inductive learning, clustering）. [Result] The proposed method raises efficiency of land use and land reclaim. Based on the SDMKD, the characteristics of TM showed that the information after using rough set is more intensive than that of none. Especially, much better results can be gained while kinds of corps are less than five. [Conclusion] This study laid significant basis for further research on data mining in the growth of crops.
