Quad-PolSAR data classification using modified random forest algorithms to map halophytic plants in arid areas.
International Journal of Applied Earth Observation and Geoinformation(2018)
摘要
•ALOS PALSAR data are used to exploit the physical properties of halophyte plant scattering mechanisms in saline wetland environments.•Two novel classifiers the random M5 model forest (RM5MF) and the classification via random forest regression (CVRFR) are proposed.•Suitability of various polarization signatures and features are evaluated for halophyte plant mapping.•The results with two study regions demonstrates substantial performances of the proposed RM5MF and CVRFR.
更多查看译文
关键词
Polarization signature,Random M5 model forest,Classification via random forest regression,Halophytic plants,ALOS-2,PolSAR image classification
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络