基于深度学习的SKA天文图像重建方法设计  

Design of SKA Astronomical Image Reconstruction Method Based on Deep Learning

作　　者：谭江龙 王合彬 徐振刚 项兆佳 张利 TAN Jianglong;WANG Hebin;XU Zhengang;XIANG Zhaojia;ZHANG Li(School of Big Data and Information Engineering,Guizhou University,Guizhou 550025,China)

机构地区：[1]贵州大学大数据与信息工程学院,贵州550025

出　　处：《集成电路应用》2024年第11期74-75,共2页Application of IC

摘　　要：阐述SKA1-LOW等现代望远镜阵列拥有极高的灵敏度、分辨率,能够观测到暗弱结构。但空间傅立叶空间不完全覆盖,暗弱结构会被psf淹没。为此,提出一种基于深度学习的天文图像复原方案,通过优化训练后能够从退化图像中恢复图像,相较于传统CLEAN算法有较大提升。This paper expounds that modern telescope arrays such as SKA1-LOW have extremely high sensitivity and resolution,and can observe dark and weak structures.But the spatial Fourier space is not completely covered,and weak structures will be submerged by PSF.It proposes a deep learning based astronomical image restoration scheme that can restore images from degraded images through optimized training,which greatly improves compared to traditional CLEAN algorithms.

关 键 词：深度学习 SKA 射电天文成像 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP394.41[自动化与计算机技术—控制科学与工程] P161[天文地球—天文学]