26 July 2007 Identification of coastal wetland using rule inferring
Author Affiliations +
The main purpose of this paper was to explore the potential of decision tree classifier in the identification and change monitoring of coastal wetland. A part of the coastal wetland in the Northern Jiangsu Province was taken as test area. Decision tree classifiers derived using different ways were applied to the classification of coastal wetland and the results were compared by independent sampling points. It was shown that the post-classification improvements by using knowledge rules could achieve higher accuracy than automatic machine learning method. In the post-classification improvements, the selection of feature vectors was crucial to the improvement of accuracy of classification.
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Renzong Ruan, Renzong Ruan, Landi Xia, Landi Xia, Ziqi Yan, Ziqi Yan, } "Identification of coastal wetland using rule inferring", Proc. SPIE 6752, Geoinformatics 2007: Remotely Sensed Data and Information, 67522L (26 July 2007); doi: 10.1117/12.760766; https://doi.org/10.1117/12.760766

Back to Top