A novel supervised classification scheme based on Adaboost for Polarimetric SAR

In this paper, a novel scheme for supervised classification problem of Polarimetric SAR images is proposed, which is based on Adaboost. Compared to traditional classifiers such as complex Wishart distribution based maximum likelihood classifier or Neural Network based classifier, the proposed method is more robust and flexible. Different features or parameters extracted from Polarimetric SAR data could be adopted into the scheme and a quantitative analysis on the significance of each parameter for classification could be achieved. Experiment results demonstrated the effectiveness of the proposed scheme.