Assessing integration of intensity, polarimetric scattering, interferometric coherence and spatial texture metrics in PALSAR-derived land cover classification

摘要: Abstract Synthetic aperture radar (SAR) is an important alternative to optical remote sensing due its ability acquire data regardless of weather conditions and day/night cycle. The Phased Array type L-band SAR (PALSAR) onboard the Advanced Land Observing Satellite (ALOS) provided new opportunities for vegetation land cover mapping. Most previous studies employing PALSAR investigated use one or two feature types (e.g. intensity, coherence); however, little effort has been devoted assessing simultaneous integration multiple features. In this study, we bridged gap by evaluating potential using numerous metrics expressing four types: polarimetric scattering, interferometric coherence spatial texture. Our case study was conducted in Central New York State, USA multitemporal imagery from 2010. classification implemented ensemble learning algorithm, namely random forest. Accuracies each classified map produced different combinations features were assessed on a pixel-by-pixel basis validation obtained stratified sample. Among evaluated, intensity most indispensable because included all highest accuracy scenarios. However, relative only metrics, combining increased overall 7%. Producer’s user’s accuracies classes improved considerably best performing combination when compared classifications single type.