Search

The segmented mirror co-phase error identification technique based on supervised learning methods has the advantages of simple application conditions, no dependence on custom sensors, a fast calculation speed, and low computing power requirements compared with other methods. However, it is often difficult to obtain a high accuracy in practical application situations with this method because of the difference between the training model and the actual model. The reinforcement learning algorithm does not need to model the real system when operating the system. However, it still retains the advantages of supervised learning. Thus, in this paper, we placed a mask on the pupil plane of the segmented telescope optical system. Moreover, based on the wide spectrum, point spread function, and modulation transfer function of the optical system and deep reinforcement learning—without modeling the optical system—a large-range and high-precision piston error automatic co-phase method with multiple-submirror parallelization was proposed. Finally, we carried out relevant simulation experiments, and the results indicate that the method is effective. [ABSTRACT FROM AUTHOR]

The wave-front phase expanded on the Zernike polynomials is estimated from a pair of images by the use of a maximum-likelihood approach, the in-focus image and the defocus image, which contaminated by noise, will greatly reduce the solution accuracy of the phase diversity (PD) algorithm. In the study, we introduce the deep denoising convolutional neural networks (DnCNNs) into the image preprocessing of PD to denoise the in-focus image and defocus the image containing gaussian white noise to improve the robustness of PD to noise. The simulation results show that the composite PD algorithm with DnCNNs is better than the traditional PD algorithm in both RMSE of phase estimation and SSIM, and the mean of the RMSE of the phase estimation of the improved PD algorithm is reduced by 78.48%, 82.35%, 71.09% and 73.67% compared with the mean of the RMSE of the phase estimation of the traditional PD algorithm. The well-trained DnCNNs runs fast, which does not increase the running time of traditional PD algorithms, and the compound approach may be widely used in various domains, such as the measurements of intrinsic aberrations in optical systems and compensations for atmospheric turbulence.

In the cophasing of the segmented optical mirrors, the Shack-Hartmann wavefront sensor is not sensitive to the submirror piston error and the large range piston errors beyond the cophasing detection range of phase diversity algorithm. It is necessary to introduce specific sensors (e.g., microlenses or prisms), but they greatly increase the complexity and manufacturing cost of the optical system. In this Letter, we introduce the convolutional neural network (CNN) to distinguish the piston error range of each submirror. To get rid of the dependence of the CNN dataset on the imaging target, we construct the feature vector by the in-focal and defocused images. The method surpasses the fundamental limit of the detection range by using different wavelengths. Finally, the results of the simulation experiment indicate that the method is effective.

Parallax observations from staggered charge-coupled devices (CCDs) have been applied to satellite jitter detection. Nevertheless, the jitter during the initial period of an imaging process cannot be detected. This paper presents an approach that combines parallax observations with the attitude data from attitude-measuring sensors in order to detect the global jitter, including the initial jitter. Low-frequency components, which can be reconstructed from attitude data, account for most jitter energy, and determine the jitter curve's overall shape. We introduce attitude data into parallax observations to constrain the initial jitter and find its optimum estimate. Meanwhile, an offset is extracted from parallax observation images by using a comprehensive matching method. A mathematical model is developed to demonstrate how to calculate the global jitter with the initial jitter and offset. Numerical simulation results indicate that, for pixel-level offset error, the root-mean-square error (RMSE) of the proposed method is 1.4 pixels, while the measurement error near integer multiples of characteristic frequency is amplified significantly. Experiments performed on Chinese Heavenly Palace-1 satellite show that the jitter at 0.12Hz with an amplitude about 6 pixels exists in the cross-track direction, while the down-track jitter results fail to show obvious periodicity.