The study of Combination Texture Information andKnowledge Base Classification for Urban Paddy AreaExtraction—Using High Resolution Satellite Image

摘要

This research uses high-resolution satellite imagesrnas a source of collecting farmland information. For effectivelyrnextract the paddy area, we use texture information and differentrnclassify methods to assist the satellite image classification.rnFirst, using maximum likelihood classifier to extract paddyrninformation form images. The results show that the UserrnAccuracy and Procedure Accuracy of the paddy area canrnincrease from 80.60% to 95.45% and 84.38% to 95.45%.rnSecond, establishing a paddy Knowledge Base and usingrnKnowledge Base Classifier to extract paddy area, and resultrnshows the User Accuracy and Producer Accuracy to be 92.16%rnand 90.06%. Finally, The result shows we can effectivelyrncontribute to the paddy field information extraction fromrnhigh-resolution satellite images.