...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>IEEE Transactions on Geoscience and Remote Sensing >Improved POLSAR Model-Based Decomposition Interpretation Under Scintillation Conditions
【24h】

Improved POLSAR Model-Based Decomposition Interpretation Under Scintillation Conditions

机译：闪烁条件下基于改进POLSAR模型的分解解释

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Low-frequency synthetic aperture radar (SAR) sensors are prone to ionospheric irregularity structures that affect the amplitude and/or phase of the radar signal. Azimuthal striping caused by amplitude scintillation in SAR data is an initial observation of such effects. In the absence of any physical model and/or technique to mitigate scintillation stripes, we have applied a 2-D fast Fourier transform (FFT) approach for improved interpretation and target identification in fully polarimetric SAR (POLSAR) model-based decomposition scattering powers. Few scenes of fully polarimetric Advanced Land Observation Satellite-Phased Array type L-band Synthetic Aperture Radar (ALOS/PALSAR) and ALOS-2/PALSAR-2 under different ionospheric conditions are acquired and used in this study. As a reference, data sets over the same areas are also acquired on different dates with negligible ionospheric activity (ionospheric quiet day). All the data sets are corrected for the Faraday rotation angle (Omega), which is estimated using the Bickel-Bates approach. Three correction strategies of scintillation-affected model-based scattering power decomposition images are presented. The strategies (zero masking, thresholding, and averaging-post-thresholding) are implemented on 2-D FFT of scattering power images. An intercomparison of corrected decomposition results from the correction strategies demonstrates the capability of 2-D-FFT-based correction strategies to improve the dominant scattering component by 3%-5% for homogeneous terrains. The correction method based on averaging-post-thresholding gives the best results that are further tested by performing a supervised classification. The overall accuracy and kappa coefficient (O A, (k) over cap) of the averaging-post-thresholding technique (72.87%, 0.59) is comparable to those of the reference data (78.74%, 0.69).
机译：低频合成孔径雷达（SAR）传感器容易受到电离层不规则结构的影响，从而影响雷达信号的幅度和/或相位。 SAR数据中的幅度闪烁引起的方位角条带化是这种影响的初步观察结果。在缺乏减轻闪烁条纹的任何物理模型和/或技术的情况下，我们已应用二维快速傅里叶变换（FFT）方法在基于全极化SAR（POLSAR）模型的分解散射功率中改进了解释和目标识别。在本研究中，很少获得和使用不同电离层条件下的全极化高级陆地观测卫星相控阵型L波段合成孔径雷达（ALOS / PALSAR）和ALOS-2 / PALSAR-2场景。作为参考，在不同日期的电离层活动（电离层安静日）也可以在不同日期获取同一区域的数据集。所有数据集都针对法拉第旋转角度（Omega）进行了校正，该角度是使用Bickel-Bates方法估算的。提出了基于闪烁的基于模型的散射功率分解图像的三种校正策略。这些策略（零掩蔽，阈值和平均后阈值）是在散射功率图像的二维FFT上实现的。来自校正策略的校正分解结果的比较表明，基于2-D-FFT的校正策略能够将均匀地形的主散射分量提高3％-5％。基于平均后阈值的校正方法可提供最佳结果，该结果可通过执行监督分类进一步测试。阈值后平均技术的整体准确性和卡伯系数（上限为O A，（k））（72.87％，0.59）与参考数据的整体准确性和kappa系数（78.74％，0.69）相当。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号