Fusion of high spatial resolution optical and polarimetric SAR images for urban land cover classification

In this paper, we proposed a new strategy for urban land cover classification by fusing high spatial resolution optical images and polarimetric SAR images. The novelty of strategy was in twofold: introduce an object-based classification method to alleviate the negative impact of inner class variability of high spatial resolution images in urban areas and geometric differences between the SAR and optical image; construct a fuzzy model to fuse features with different properties and ranges. Experiments were carried out to validate the strategy by using eCognition8.0. 1m RGB airborne images and 4m quad-polarization RADARSAT-2 images in Zhangye City, Gansu Province were used. Both of the images were acquired in July 2012. Results indicated that the optical image shows good performance at extracting land covers with distint spectral features such as natural vegetation, while the SAR image is better at differentiating several other land covers with similar spectral identities. For example, the introduction of polarimetric SAR features clearly improve the classification accuracy of bare soil and buildings by about 5%-10%, and also help the separation of artificial and natural vegetation to some extent. The overall classification accuracy increases from 85% to 88.18%. Based on object-based classification and fuzzy deduction, the proposed strategy proves a promising tool in fusing SAR and optical images to extract urban land cover.