首页 >  , Vol. , Issue () : -

摘要

全文摘要次数: 311 全文下载次数: 572
引用本文:

DOI:

10.11834/jrs.20233002

收稿日期:

2023-01-01

修改日期:

2023-03-14

PDF Free   EndNote   BibTeX
基于多尺度注意力机制的 PolSAR深度学习超分辨率模型
林镠鹏1, 李杰2, 沈焕锋1
1.武汉大学 资源与环境科学学院;2.武汉大学测绘学院
摘要:

全极化合成孔径雷达影像(PolSAR)可提供丰富的极化信息,但成像系统限制使其空间分辨率受到制约。为此,本文基于深度学习框架,提出一种基于多尺度注意力机制的超分辨率重建网络(MS-PSRN),通过对低分辨率PolSAR影像进行分辨率增强,生成高分辨率的PolSAR影像。在该模型框架下,采用多尺度注意力模块对不同尺度下的地物目标进行特征提取;提出联合式与分离式两种内嵌方式,在模型中嵌入通道注意力与空间注意力,利用注意力机制的权值重校准特性,增强PolSAR影像的极化信息与空间信息;引入残差信息蒸馏机制,提取判别性特征并对模型参数进行压缩;提出自适应损失函数对网络训练过程进行约束以提升模型的数值拟合能力以及边缘信息保持能力。本文采用RADARSAT-2卫星的模拟数据与真实数据两个数据集对提出的方法进行验证。空间信息实验结果表明,本文方法在目视结果与定量指标中均优于对比算法,具有更高的空间纹理细节重建精度与较低的重建误差。极化信息保持测试表明,本文方法可在提升空间分辨率的同时,有效保持PolSAR影像的极化信息。

PolSAR image deep learning super-resolution model based on multi-scale attention mechanism
Abstract:

Fully polarimetric synthetic aperture radar imagery (PolSAR) can provide rich polarimetric information, however, due to the limitations of the system"s signal bandwidth and the physical size of the antenna, the spatial resolution of the SAR imaging system is restricted while acquiring multiple polarization information. To solve this problem, based on the deep learning framework, this paper proposes a multi-scale attention-based PolSAR super-resolution network (MS-PSRN), which performs super-resolution reconstruction on the low-resolution full-polarimetric SAR images to generate the fully polarimetric SAR images with high spatial resolution. Under this super-resolution reconstruction framework, this paper uses a multi-scale architecture to fully extract the feature information of objects at different scales. On this basis, the spatial attention mechanism and the channel attention mechanism are introduced to recalibrate the feature maps, which is used to enhance the reconstruction performance of spatial details and improve the ability to maintain polarization information respectively. Two attention mechanism embedding methods, joint and separated, are proposed to cope with the spatial size and quantity changes of the feature maps processed by the encoder and decoder. This paper introduces a residual information distillation mechanism, extracts discriminative features through feature distillation, and compresses model parameters at the same time. In addition, the adaptive loss function is proposed to constrain the network training process and improve the model"s numerical fitting ability and edge information preservation ability. In this paper, the proposed method is verified by two data sets, which is the simulated data and the real data produced by RADARSAT-2 images. The experimental results of spatial information show that the proposed method is superior to the comparison algorithms in both visual results and quantitative indicators, and has higher texture detail reconstruction accuracy and lower reconstruction error. The polarimetric information preservation test shows that the proposed method can effectively preserve the polarimetric information of PolSAR images while improving the spatial resolution.

本文暂时没有被引用!

欢迎关注学报微信

遥感学报交流群