We propose a convolutional neural network (CNN) based method, namely phase diversity convolutional neural network (PD-CNN) for the speed acceleration of phase-diversity wavefront sensing. The PD-CNN has achieved a state-of-the-art result, with the inference speed about $0.5$ ms, while fusing the information of the focal and defocused intensity images. When compared to the traditional phase diversity (PD) algorithms, the PD-CNN is a light-weight model without complicated iterative transformation and optimization process. Experiments have been done to demonstrate the accuracy and speed of the proposed approach.

This paper proposes a supervised deep neural network model for accomplishing highly efficient image quality assessment (IQA) for adaptive optics (AO) images. The AO imaging systems based on ground-based telescopes suffer from residual atmospheric turbulence, tracking error, and photoelectric noise, which can lead to varying degrees of image degradation, making image processing challenging. Currently, assessing the quality and selecting frames of AO images depend on either traditional IQA methods or manual evaluation by experienced researchers, neither of which is entirely reliable. The proposed network is trained by leveraging the similarity between the point spread function (PSF) of the degraded image and the Airy spot as its supervised training instead of relying on the features of the degraded image itself as a quality label. This approach is reflective of the relationship between the degradation factors of the AO imaging process and the image quality and does not require the analysis of the image’s specific feature or degradation model. The simulation test data show a Spearman’s rank correlation coefficient (SRCC) of 0.97, and our method was also validated using actual acquired AO images. The experimental results indicate that our method is more accurate in evaluating AO image quality compared to traditional IQA methods.

Measuring the intensity and distribution of atmospheric optical turbulence at large-aperture astronomical telescope sites is crucial to optimizing turbulence correction for different layers. A real-time measurement of turbulence distribution in large-aperture telescopes would be valuable for the parameter optimization of adaptive optics (AO) systems, especially for large field-of-view AO systems such as multi-conjugate adaptive optics (MCAO) and ground-layer adaptive optics (GLAO). Based on the GLAO system of NVST at FSO, a real-time measurement system was deployed to assess the site’s atmospheric turbulence intensity and distribution. This system is, to our knowledge, the first real-time turbulence parameter measurement system in the world with an AO system. We adopt pseudo-open loop methods to restore the turbulence information from the close-loop data of GLAO and measure the turbulence strength and distribution. Multiple subaperture pairs are used instead of a pair of subapertures for fitting calculation to increase the measurement accuracy. Two conventional measurement algorithms, SLODAR and S-DIMM+, are compared with the data from the open-source simulator SOAPY, to cross-verify the correctness of our calculation based on the data process of pseudo-open loop data and multiple subaperture pairs. The simulation results show that for two layers’ turbulence input, approximately 93% of the turbulence is correctly detected with the SLODAR method and the given parameters of wavefront sensors and correctors, while the S-DIMM+ is 87%. Real-time measurements of atmospheric turbulence at the NVST site were carried out on 28 May 2023. The observation results indicated that approximately 80% of the turbulence was located below an altitude of 2000 m; only a few appear in the upper height.

The reconstruction of 3D shapes from a single view has been a longstanding challenge. Previous methods have primarily focused on learning either geometric features that depict overall shape contours but are insufficient for occluded regions, local features that capture details but cannot represent the complete structure, or structural features that encode part relationships but require predefined semantics. However, the fusion of geometric, local, and structural features has been lacking, leading to inaccurate reconstruction of shapes with occlusions or novel compositions. To address this issue, we propose a two-stage approach for achieving 3D shape reconstruction. In the first stage, we encode the hierarchical structure features of the 3D shape using an encoder-decoder network. In the second stage, we enhance the hierarchical structure features by fusing them with global and point features and feed the enhanced features into a signed distance function (SDF) prediction network to obtain rough SDF values. Using the camera pose, we project arbitrary 3D points in space onto different depth feature maps of the CNN and obtain their corresponding positions. Then, we concatenate the features of these corresponding positions together to form local features. These local features are also fed into the SDF prediction network to obtain fine-grained SDF values. By fusing the two sets of SDF values, we improve the accuracy of the model and enable it to reconstruct other object types with higher quality. Comparative experiments demonstrate that the proposed method outperforms state-of-the-art approaches in terms of accuracy.

Atmospheric turbulence is a critical factor limiting the imaging resolution of ground-based solar telescopes. This study presents a systematic investigation of the intensity and vertical distribution of daytime atmospheric turbulence during winter at the Fuxian Solar Observatory, using data acquired from the 1-meter New Vacuum Solar Telescope (NVST) with its Ground Layer Adaptive Optics (GLAO) system and a custom-developed wide-field Shack–Hartmann wavefront sensor. Statistical results reveal a median Fried parameter ($r_0$) of 8.25 cm at 500 nm, indicating generally favorable daytime observing conditions. A distinct diurnal variation in $r_0$ was observed: values were higher in the morning and afternoon but decreased significantly around noon due to enhanced ground-layer heating. Vertical turbulence profiling showed that approximately 52.8% of the total turbulence strength originates from the ground layer, and 93.1% is confined below 4 km, with only weak turbulence detected at higher altitudes. This study establishes the first statistical turbulence profile model for the Fuxian Solar Observatory site during winter daytime, providing crucial insights for optimizing high-resolution solar observations and the design of multi-conjugate adaptive optics systems.

This paper proposes a lucky imaging based blind deconvolution algorithm for wide field-of-view (FoV) ground layer adaptive optics (GLAO) solar images. Our method effectively combines the advantages of traditional lucky imaging and blind deconvolution techniques, by segmenting and reconstructing sub-images in different field directions of the captured image sequence, as well as stitching and fusing reconstructed results of multiple regions, which significantly improves the robustness of reconstruction results. High quality wide FoV reconstruction results were obtained for actual GLAO solar images collected by a 1 m New Vacuum Solar Telescope in experiments. After processing, the Image Sharpness Metric Based on the Notion of Just Notifiable Blur improved by 51.39% and 53.65% compared to the average frame of the input image. The algorithm proposed can help achieve good results in every field direction for wide FoV images, which can provide important data support for subsequent research.