Local and Global Spatial Information for Land Cover Semisupervised Classification of Complex Polarimetric SAR Data

Abstract

Each of the three satellites constituting the RADARSAT Constellation Mission (RCM) provide compact polarimetric synthetic aperture radar (CP SAR) data. The complex CP data have similar properties to the complex quad polarimetric (QP) data provided by prior RADARSAT missions. In this paper, a land cover classification method using spatial information is designed based on the statistical characteristics of the complex CP and QP SAR data. First, the local spatial dependency among pixels is captured by superpixels. Second, a graph is constructed on the superpixels to model the global spatial dependency among superpixels. The land cover classification image with land cover type labels is then estimated by propagating labels from the few labeled superpixels to the unlabeled superpixels. Classification of two RCM complex CP and QP scenes demonstrates that the proposed method, with few labeled pixels, provides much higher classification accuracy than methods which do not exploit global spatial dependency.

BSDSNet: Dual-Stream Feature Extraction Network Based on Segment Anything Model for Synthetic Aperture Radar Land Cover Classification

“…Existing PolSAR land cover classification methods can be categorized into traditional approaches [5,6] and deep learning methods [7][8][9]. Traditional classification methods are predominantly designed based on statistical features of PolSAR data, such as the Wishart distribution [10], etc.…”

Synthetic Aperture Radar in Vineyard Monitoring: Examples, Demonstrations, and Future Perspectives

“…In fact, the use of SAR in viticulture can already be found in several studies. The use of space-borne SAR in viticulture was reported in land cover classifications [11][12][13], soil moisture monitoring [14][15][16], and, particularly, in polarimetric applications of SAR data [17][18][19]. For example, in land cover classification, space-borne SAR can be used to differentiate between vineyard areas and other types of land cover such as forests or urban areas based on their radar backscattering characteristics.…”