A Hybrid Entropy Decomposition and Support Vector Machine Method for Agricultural Crop Type Classification
(0) This paper presents the development of Synthetic Aperture Radar (SAR) image classifier based on the hybrid method of Entropy Decomposition and Support Vector Machine (EDSVM) for agricultural crop type classification. The Support Vector Machine (SVM) is successfully applied to the key parameters extracted from Entropy Decomposition to obtain good image classifications. In this paper, this novel classifier has been applied on a multi-crop region of Flevoland, Netherlands with multi-polarization data for crop type classification. Validation of the classifiers has been carried out by comparing the classified image obtained from EDSVM classifier and SVM. The EDSVM classifier demonstrates the advantages of the valuable decomposed parameters and statistical machine learning theory in performing better results compared with the SVM classifier. The final outcome of this research clearly indicates that EDSVM has the ability in improving the classification accuracy for agricultural crop type classification.
Chue-Poh Tan Hong-Tat Ewe Hean-Teik Chuah
Faculty of Engineering, Multimedia University Persiaran Multimedia, 63100 Cyberjaya, Selangor, Malay Faculty of Information Technology, Multimedia University Persiaran Multimedia, 63100 Cyberjaya, Sela
国际会议
Progress in Electromagnetics Research Symposium 2007(2007年电磁学研究新进展学术研讨会)(PIERS 2007)
北京
英文
42-46
2007-03-26（万方平台首次上网日期，不代表论文的发表时间）