The server is under maintenance between 08:00 to 12:00 (GMT+08:00), and please visit later.
We apologize for any inconvenience caused
Login  | Sign Up  |  Oriprobe Inc. Feed
China/Asia On Demand
Journal Articles
Laws/Policies/Regulations
Companies/Products
SAR target super-resolution based on transfer learning
Author(s): 
Pages: 1940-1952
Year: Issue:  6
Journal: Acta Aeronautica ET Astronautica Sinica

Keyword:  synthetic aperture radarsuper-resolutiontransfer learningsemi-coupled dictionarysparse representation;
Abstract: 针对合成孔径雷达(SAR)目标超分辨重建问题,提出了一种基于迁移学习的超分辨方法.在光学图像梯度域中联合训练超完备字典与稀疏编码映射,利用半耦合字典联系SAR图像与光学图像,寻找SAR图像在半耦合字典下的稀疏编码,并在高分辨率字典下完成重建.结合SAR图像的先验信息,使用正则化方法对SAR目标进行特征增强.所提方法在TerraSAR-X数据和MSTAR数据上进行了仿真实验,重建结果表明,相比目前的插值方法和稀疏表示方法,所提方法空间分辨率可提高0.5~1.5个像素.正则化增强结果表明,引入稀疏先验的正则化增强能够进一步提高空间分辨率并抑制杂波比,最后分析了正则化参数的选取对图像质量的影响.
Related Articles
loading...