Research on scattering characteristics of buildings under trees in polarimetric SAR images based on polar decomposition

Accurate detection of buildings is of great significance for urban planning and rapid development. In southern regions, due to climatic and geographical factors, vegetation is dense throughout the year, and low-rise buildings are easily obscured by trees, leading to misdetections and omissions. Polarimetric SAR images contain rich information about land features and record the polarization scattering information of target objects. Utilizing the penetrating capability of long-wavelength SAR, a de-orienting Yamaguchi decomposition is applied to buildings obscured by trees. The surface scattering, double-bounce scattering, and volume scattering components are then used to analyze the scattering characteristics of buildings under trees. An experimental analysis was conducted using ALOS PALSAR imagery of Wuhan city from August 16, 2019. The results indicate that buildings obscured by trees exhibit different scattering characteristics compared to other buildings, and the analysis of their scattering component information holds great potential for land classification and building detection.