Data underlying the publication: RCNet: A Hybrid Framework for PolSAR Image Classification via Real-Complex Polarimetric Feature Fusion

Three benchmark PolSAR datasets were utilized to assess the performance of the proposed method in this study. Oberpfaffenhofen, San Francisco and Barnaul. The first dataset was acquired by the L-band PolSAR data from the E-SAR sensor, capturing the Oberpfaffenhofen area in Germany with a size of 1200 × 1300 pixels, and spatial resolution is 3 × 3 meters. The primary land cover types in this region include dense urban areas, forests, and some regions without artificial markings. The secondary dataset was collected via the airborne L-band PolSAR sensor aboard NASA/JPL's AIRSAR platform, covering the San Francisco region This dataset exhibits a spatial extent of 1024 × 900 pixels with 10-meter resolution, encompassing diverse land cover categories including urban zones, vegetated areas, rocky terrain, oceanic surfaces, and exposed landforms.The third dataset is from China’s Gaofen-3 satellitecollected on May 5, 2018. The multi-look data consists of 1474 × 1310 pixels. The observed area includes various land cover types such as urban areas, forests, farmland.These datasets are provided in this repository.

本研究采用三组基准极化合成孔径雷达（PolSAR）数据集以评估所提方法的性能，对应区域分别为奥伯法芬霍芬（Oberpfaffenhofen）、旧金山（San Francisco）与巴尔瑙尔（Barnaul）。第一组数据集由E-SAR传感器获取的L波段PolSAR数据构成，采集范围为德国奥伯法芬霍芬区域，图像尺寸为1200×1300像素，空间分辨率为3×3米。该区域主要土地覆盖类型包括高密度城区、森林以及部分无人工标识的区域。第二组数据集由搭载于美国国家航空航天局/喷气推进实验室（NASA/JPL）AIRSAR平台的机载L波段PolSAR传感器采集得到，覆盖旧金山区域。该数据集空间尺寸为1024×900像素，分辨率为10米，涵盖城区、植被区、岩石地形、海洋表面与裸露地貌等多种土地覆盖类别。第三组数据集来自中国高分三号（Gaofen-3）卫星，采集于2018年5月5日，为多视处理数据，尺寸为1474×1310像素。观测区域包含城区、森林与农田等多种土地覆盖类型。本仓库已提供上述全部数据集。