Optical systems suffer from wavefront aberrations due to complex atmospheric environments and system component errors, leading to systematic aberrations and significantly degrading optical field quality. Therefore, the detection and correction of optical aberrations are crucial for efficient and accurate observations. To fully utilize the capabilities of observation equipment and achieve high-efficiency, accurate imaging, it is essential to develop wavefront correction technologies that enable ultra-precise wavefront control. The application of data-driven techniques in wavefront correction can effectively enhance correction performance and better address complex environmental challenges. This paper elaborates on the research progress of data-driven methods in wavefront correction from three aspects: principles, current research status, and practical applications. It analyzes the performance of data-driven methods in diverse real-world scenarios and discusses future trends in the deep integration of data-driven approaches with optical technologies. This work provides valuable guidance for advancing wavefront correction methodologies.

Wavefront reconstruction and correction under strong turbulence has become a crucial issue for adaptive optics systems. An approach using a plenoptic sensor has attracted high attention recently. However, the existing reconstruction and correction algorithms based on the plenoptic sensor are sometimes unsatisfactory. Therefore, we propose a modified approach based on the checkerboard algorithm to reconstruct the wavefront. By means of a weighted summation, the reconstruction accuracy is clearly increased. We propose using the direct-gradient method to correct the disturbed wavefront after reconstruction, and our experimental results demonstrate the effectiveness with only one correction step.

Single-pixel imaging is a computational optical imaging technique that uses a single-pixel detector to obtain scene information and reconstruct the image. Compared with traditional imaging techniques, single-pixel imaging has the advantages of high sensitivity and a wide dynamic range, etc., which make it have broad application prospects in special frequency band imaging and scattering media imaging. This paper mainly introduces the history of development and the characteristics of the single-pixel detector, focuses on the typical applications of single-pixel imaging in coded aperture, transverse scanning, and longitudinal scanning systems, and gives an account of the application of deep learning technology in single-pixel imaging. At the end of this paper, the development of single-pixel imaging is summarized and future trends forecasted.

To improve the quality of seismic data acquired by distributed acoustic sensing (DAS) systems and mitigate phase noise in phase-sensitive optical time domain reflectometry (Φ-OTDR) systems, a denoising method based on successive variational mode decomposition (SVMD) and the shifting window Pearson correlation coefficient (SWPCC), termed SVMD-SWPCC, is proposed. The phases obtained by the Φ-OTDR systems are adaptively decomposed into several intrinsic mode functions (IMFs) using SVMD. The proposed SWPCC method is used to accurately identify highly correlated components from each IMF, enhancing the overall signal-to-noise ratio (SNR). The SVMD-SWPCC algorithm improves upon existing approaches that determine which modes need to be retained on the basis of global correlation coefficients. By introducing a shifting window, the impact of unrelated components within IMFs on the Pearson correlation coefficient (PCC) is mitigated, thereby significantly improving the accuracy of the correlation assessment. The experimental results for single-frequency, multi-frequency, sweep-frequency, and wide-frequency impact signals show that the SVMD-SWPCC algorithm effectively mitigates the phase noise in Φ-OTDR. Compared to existing methods, SVMD-SWPCC offers advantages such as an improved SNR and robust adaptability. It has also been successfully applied to denoising in controlled electromagnetic seismic experiments and sledgehammer impact signal tests.

Spatial multispectral imaging technology can selectively image in specific spectral bands, and the filter wheel is a core component for multispectral selection. At present, there are relatively few types of spectral bands for the filter wheel under limited space/weight constraints. Addressing the challenges presented by this issue, this paper introduces an innovative design approach for the development of a double-layer or even multi-layer filter wheel that is operated by a solitary motor in conjunction with a differential gear mechanism, enabling a vast array of spectral segment combinations within a highly compact layout. A detailed design is implemented for the double-layer filter wheel, including comprehensive modal and dynamic analyses. The results of the modal analysis attested to the structural stability of the component, and the outcomes of the dynamic analysis validated the component’s timely and reliable switching capabilities. A prototype was meticulously crafted and subjected to rigorous testing. The switching functionality was validated during these tests, concurrently affirming the accuracy of the finite element analysis results. Additionally, spectral and application testing confirmed the number of spectral segments and the practical utility of the components. The research presented in this article introduces an innovative design concept for multispectral imaging filter-wheel mechanisms, providing a valuable reference and profound insights for the design and arrangement of a double-layer or even multi-layer filter wheel.