Multitemporal polarimetric RADARSAT-2 SAR data for urban land cover mapping through a dictionary-based and a rule-based model selection in a contextual SEM algorithm

摘要

This paper presents a dictionary- and rule-based model selection approach in an adaptive contextual semisupervisednalgorithm for improving urban land cover classification using high-resolution multitemporal RADARSAT-2npolarimetric SAR (PolSAR) data. Six-date PolSAR data were acquired from June to September, 2008, over the GreaternToronto Area. Contextual information and the capabilities of different PolSAR distribution models were explored by thenspatially variant Finite Mixture Model (FMM) with an adaptive Markov Random Field (MRF) in a StochasticnExpectationu0002Maximization (SEM) algorithm. This algorithm can obtain homogenous results while preserving shapendetails in the complex urban environment with high accuracy. Commonly used PolSAR distribution models such asnWishart, G0p, Kp, and KummerU were compared through the proposed approaches for urban land cover mapping.nAccording to a Goodness-of-Fit test based on Mellin transformation, an accurate PolSAR distribution model could benselected with the dictionary-based classification. However, the results showed that improvement from the dictionary-basednapproach was limited. Therefore, further improvements were expected by exploring expert knowledge. The initial resultsnshowed that G0p and KummerU performed better for distinguishing between low density built-up areas and forest. G0p,nKp, and KummerU are better for the low scattering classes. The Wishart model has superior capacity in separating highndensity built-up areas and the adjacent roads. Based on such knowledge, a set of rules was developed to integrate thenadvantages of alternative models. Significant improvement on the overall classification accuracy could be observed by thisnrule-based approach. The biggest improvement was achieved using the HDu0002Road rule on the G0p model with the bestnoverall classification accuracy at 89.99% (kappa: 0.87). This represented 4.1% (kappa: 0.045) improvement over that ofnG0p without model selection.