SAR_Despeckling

The dataset consists of Sentinel-1 SAR data sampled from areas near Shanghai, China, capturing diverse urban and natural landscapes. The polarization mode employed is VV+VH, providing both vertical-vertical and vertical-horizontal polarized signals to enhance the richness of the information. To simulate real-world conditions, gamma noise has been intentionally introduced, mimicking the noise that typically arises during SAR image sampling and processing. This addition ensures that the dataset is representative of actual operational scenarios, making it suitable for testing and validating various SAR image denoising and analysis algorithms. The dataset serves as a valuable resource for researchers focused on SAR image despeckling, feature extraction, and interpretation.

本数据集包含采集自中国上海周边区域的哨兵-1号（Sentinel-1）合成孔径雷达（Synthetic Aperture Radar，SAR）数据，覆盖多样化的城市与自然地貌景观。该数据集采用VV+VH极化模式，可同时获取垂直-垂直（VV）与垂直-水平（VH）极化信号，有效提升信息丰富度。为模拟真实应用场景，本数据集刻意添加了伽马噪声（Gamma noise），以复现合成孔径雷达图像在采集与处理过程中常见的噪声特征。该噪声添加操作确保数据集能够反映真实的业务场景，适用于各类SAR图像去噪与分析算法的测试与验证工作。本数据集可为聚焦于SAR图像斑点抑制（despeckling）、特征提取与图像解译的研究人员提供宝贵的研究资源。