Dissertation
Dissertation > Astronomy,Earth Sciences > Surveying and Mapping > Photogrammetry and Surveying, Mapping and Remote Sensing > Surveying, Mapping and Remote Sensing technology

The Classification of Remote Sensing Image Based on Spatial Data Mining and Knowledge Discovery

Author ChenXiaoZuo
Tutor YuMing
School Fujian Normal University
Course Cartography and Geographic Information Systems
Keywords Remote sensing image classification Spatial Data Mining and Knowledge Discovery C4.5
CLC P237
Type Master's thesis
Year 2007
Downloads 395
Quotes 8
Download Dissertation

The use of remote sensing technology to investigate the regional land-use types , quantitative extraction of land use information is currently one of the important research areas . Remote sensing image classification of remote sensing data in the first step of the analysis and application of land resources , how to solve the multi-class image recognition and meet certain accuracy is a key issue in the study of remote sensing images , has a very important significance . More complex feature type Fuzhou Outskirts area a small area of land use classification , for example, spectral characteristics, texture characteristics and topographical features of the integration of remote sensing images to build a multi-source spatial database using C4.5 algorithm training sample data set of spatial database found that the classification rules to classify experimental comparison and analysis with the traditional supervised classification and logical channel classification . The results show that the classification accuracy than traditional supervised classification and logical channel classification method based on C4.5 classification algorithm . C4.5 algorithm to build a decision tree to obtain the classification rules is therefore reasonable , it can quickly and efficiently for a large number of classification rules , it is an effective means to promote knowledge - based remote sensing image classification method is widely used in the land use classification .

Related Dissertations
More Dissertations