Real-time, multiframe, blind deconvolution of solar images

被引:36
作者
Asensio Ramos, A. [1 ,2 ]
Rodriguez, J. de la Cruz [3 ]
Pastor Yabar, A. [1 ,2 ,4 ]
机构
[1] Inst Astrofis Canarias, Tenerife 38205, Spain
[2] Univ La Laguna, Dept Astrofis, Tenerife 38205, Spain
[3] Stockholm Univ, Albanova Univ Ctr, Inst Solar Phys, Dept Astron, S-10691 Stockholm, Sweden
[4] Kiepenheuer Inst Sonnenphys, D-79104 Freiburg, Germany
基金
欧洲研究理事会; 瑞典研究理事会;
关键词
techniques: image processing; methods: data analysis; Sun: atmosphere; RESTORATION; RESOLUTION;
D O I
10.1051/0004-6361/201833648
中图分类号
P1 [天文学];
学科分类号
0704 ;
摘要
The quality of images of the Sun obtained from the ground are severely limited by the perturbing effect of the Earth's turbulent atmosphere. The post-facto correction of the images to compensate for the presence of the atmosphere require the combination of high-order adaptive optics techniques, fast measurements to freeze the turbulent atmosphere, and very time-consuming blind deconvolution algorithms. Under mild seeing conditions, blind deconvolution algorithms can produce images of astonishing quality. They can be very competitive with those obtained from space, with the huge advantage of the flexibility of the instrumentation thanks to the direct access to the telescope. In this contribution we make use of deep learning techniques to significantly accelerate the blind deconvolution process and produce corrected images at a peak rate of similar to 100 images per second. We present two different architectures that produce excellent image corrections with noise suppression while maintaining the photometric properties of the images. As a consequence, polarimetric signals can be obtained with standard polarimetric modulation without any significant artifact. With the expected improvements in computer hardware and algorithms, we anticipate that on-site real-time correction of solar images will be possible in the near future.
引用
收藏
页数:16
相关论文
共 28 条
[1]  
[Anonymous], 2010, INT C MACH LEARN
[2]   DeepVel : Deep learning for the estimation of horizontal velocities at the solar surface [J].
Asensio Ramos, A. ;
Requerey, I. S. ;
Vitas, N. .
ASTRONOMY & ASTROPHYSICS, 2017, 604
[3]   ANALYSIS OF SEEING-INDUCED POLARIZATION CROSS-TALK AND MODULATION SCHEME PERFORMANCE [J].
Casini, R. ;
de Wijn, A. G. ;
Judge, P. G. .
ASTROPHYSICAL JOURNAL, 2012, 757 (01)
[4]   Enhancing SDO/HMI images using deep learning [J].
Diaz Baso, C. J. ;
Asensio Ramos, A. .
ASTRONOMY & ASTROPHYSICS, 2018, 614
[5]   Deep Residual Learning for Image Recognition [J].
He, Kaiming ;
Zhang, Xiangyu ;
Ren, Shaoqing ;
Sun, Jian .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :770-778
[6]   A CATALOG OF VISUAL-LIKE MORPHOLOGIES IN THE 5 CANDELS FIELDS USING DEEP LEARNING [J].
Huertas-Company, M. ;
Gravet, R. ;
Cabrera-Vives, G. ;
Perez-Gonzalez, P. G. ;
Kartaltepe, J. S. ;
Barro, G. ;
Bernardi, M. ;
Mei, S. ;
Shankar, F. ;
Dimauro, P. ;
Bell, E. F. ;
Kocevski, D. ;
Koo, D. C. ;
Faber, S. M. ;
Mcintosh, D. H. .
ASTROPHYSICAL JOURNAL SUPPLEMENT SERIES, 2015, 221 (01)
[7]  
Ioffe S, 2015, PR MACH LEARN RES, V37, P448
[8]  
Kingma D. P., P 3 INT C LEARN REPR
[9]  
LABEYRIE A, 1970, ASTRON ASTROPHYS, V6, P85
[10]  
Ledig C., 2017, P CVPR, V2, P4