Deep learning wavefront sensing for fine phasing of segmented mirrors

Abstract

Segmented primary mirror provides many crucial important advantages for the construction of extra-large space telescopes. The imaging quality of this class of telescope is susceptible to phasing error between primary mirror segments. Deep learning has been widely applied in the field of optical imaging and wavefront sensing, including phasing segmented mirrors. Compared to other image-based phasing techniques, such as phase retrieval and phase diversity, deep learning has the advantage of high efficiency and free of stagnation problem. However, at present deep learning methods are mainly applied to coarse phasing and used to estimate piston error between segments. In this paper, deep Bi-GRU neural work is introduced to fine phasing of segmented mirrors, which not only has a much simpler structure than CNN or LSTM network, but also can effectively solve the gradient vanishing problem in training due to long term dependencies. By incorporating phasing errors (piston and tip-tilt errors), some low-order aberrations as well as other practical considerations, Bi-GRU neural work can effectively be used for fine phasing of segmented mirrors. Simulations and real experiments are used to demonstrate the accuracy and effectiveness of the proposed methods.

Artificial Intelligence in Astronomical Optical Telescopes: Present Status and Future Perspectives

“…A DNN can be used to directly construct the relationship between a point map obtained by a Shack-Hartmann wavefront sensor and the calibration voltage to improve calibration efficiency [63]. Bi-GRU can be used to obtain the corresponding relationship between the defocused star images and the wavefront [64]. A sketch map of the co-phasing approach using the Bi-GRU network is shown in Figure 5.…”

“…Figure 5.In-focus and defocused star images and aberrated wavefront maps are used to train the Bi-GRU network, and the trained network is used to predict wavefront maps[64].…”

Nonuniform Correction of Ground-Based Optical Telescope Image Based on Conditional Generative Adversarial Network

“…The methods based on pre-calibration or iteration have problems such as being only suitable for a single scene or a large amount of calculation. In recent years, with the wide application of deep learning in various fields, optical problems are also more widely solved by deep learning, including optical interferometry [ 13 ], single-pixel imaging [ 14 , 15 ], wavefront sensing [ 16 , 17 , 18 ], remote sensing [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ] and Fourier ptychography [ 27 , 28 , 29 ]. Using deep learning for image enhancement in imaging systems is also more attractive [ 22 , 23 , 30 , 31 , 32 ].…”

Piston Sensing for Golay-6 Sparse Aperture System with Double-Defocused Sharpness Metrics via ResNet-34

“…Xiafei Ma et al demonstrated that using only a single deep convolutional neural network (DCNN) is sufficient to detect pistons from a broadband extended image [ 10 ]. Yirui Wang et al authenticated that a Bi-GRU neural work with a much simpler structure can be effectively used for delicate phase segmented mirrors [ 11 ].…”