Search

The development of communication technologies has led to an ever-increasing demand for higher dimension and higher speed of optical coding technology. The orbital angular momentum (OAM) of vortex beams provides a new degree of freedom for optical coding in recent years. In this paper, we propose a novel high-dimensional vortex beam encoding/decoding scheme for high-speed free-space communication. An off-axis 4f-configuration with a digital micromirror device (DMD) and a MIP-based pattern formation algorithm are employed to generate fast switchable, accurate vortex beams (arrays) that can be used for encoding. Interference is introduced to efficiently decode data information. We experimentally demonstrate a high-quality communication link based on 64-ary vortex beam encoding/decoding and zero bit error rate is observed, which shows the validity of the proposed high-dimensional scheme. Owing to enhancement of encoding dimension and employment of high-speed DMDs, the proposed scheme enjoys a higher transfer rate exceeding other communication technologies. What is more, the data transfer rate can be further improved by introducing vortex arrays, which is demonstrated by the rapid transmission of a 24-bit true color bitmap. [ABSTRACT FROM AUTHOR]

Abstract A clear image of observed object may deteriorate into an unrecognizable speckle pattern when encountering with heterogeneous scattering media, thus it is necessary to recover the object image from the speckle pattern. Here, a machine-learning-based support vector regression (SVR) method for imaging through scattering media is experimentally demonstrated. The proposed method learns inverse scattering function (ISF) with known object-and-speckle pairs, then reconstructs unknown object with the learned ISF. Essential normalization preprocessing is pre-performed before learning the ISF. Experiments show that more training pairs lead to more accurate ISF and higher reconstruction fidelity. The proposed method provides a general solution for imaging through scattering media and is expected to has its potential applications on inverse problems, such as phase retrieval. Highlights • A machine-learning based support vector regression method for imaging through scattering media is proposed. • The method is experimentally validated to be data-driven and feature-driven, as well as noise robust to some extent. • The method realizes inverse scattering imaging, and has its potential in inverse problems such as phase retrieval. [ABSTRACT FROM AUTHOR]

Control over both amplitude and phase of optical beams plays an important role in optics, especially in the field of optical trapping. Trapping particles needs optical beams with specified intensity distribution and desired phase distribution, so joint control over light beams with high precision is indispensable. In this paper, a novel joint amplitude and phase control method specially designed for high precision using a digital micromirror device (DMD) is proposed. An off-axis 4f-configuration and a pattern formation algorithm are required for the implementation of the proposed method. Getting DMD pattern by considering the correlations among all pixels guarantees the high precision but increases the computation complexity, which makes it difficult to realize. Thus we present a pattern formation algorithm based on mixed integer programming and its improved version with higher computing efficiency. Experimental tests have been conducted to verify the superior performance of the proposed method over the conventional methods, such as Lee holography and Superpixel method. Measured results and quantitative analyses show that the proposed method enables simultaneous and independent control over amplitude and phase of light beams with a higher precision and performs better when steepest phase gradients exist in the target field. What is more, vortex beam and line beam can be generated accurately by the proposed method. [ABSTRACT FROM AUTHOR]

For conventional multichannel synthetic aperture radar (SAR) systems, the pulse repetition frequency is greater than the azimuth Doppler bandwidth. Therefore, each receive channel can obtain unambiguous SAR image and the moving target usually overlaps with no more than one clutter. However, for high‐resolution wide‐swath (HRWS) SAR systems, the PRF is usually much less than the Doppler bandwidth and thus the moving target may overlap with quite a few clutters, which will severely affect the estimation of the radial velocity of the moving target. A novel radial velocity estimation method based on maximum likelihood estimation (MLE) for HRWS SAR systems is presented. Even if the moving target overlaps with quite a few clutters, the proposed MLE method can still effectively estimate the radial velocity. In order to reduce the computational load, the proposed method transforms the multi‐dimensional spectral peak searching of MLE into one‐dimensional peak searching and reduces the times of matrix inversion by a block matrix method. Theoretical analysis and experiments demonstrate the feasibility of the proposed method. [ABSTRACT FROM AUTHOR]

This Letter addresses the coherent integration problem for air and space moving target detection in a staggered pulse repetition interval (PRI) radar system. In this algorithm, a range‐compressed signal is firstly transformed into the one in the range frequency domain, and then the authors resample the non‐uniformly sampled signal into a uniform grid along azimuth direction. Subsequently, the well‐known Keystone transform is applied to compensate the range cell migration caused by target motion. Finally, a moving target can be well focused in the range‐Doppler domain after motion compensation. Simulated results are provided to validate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Through taking maximum advantage of fast data processing characteristic of optical computing, optronic convolutional neural networks (OPCNNs) make all computation operations executed in optics to reduce calculation burden in electronic to the maximum extent and successfully performing classification tasks on both hand-written digit dataset and fashion product dataset. However, practical performance of OPCNNs is sensitive to the position variation of testing images caused by whether the alignment error of components or the translation of classified target. The classification performance reduction in such situations severely limits the application scenarios of OPCNNs. In this paper, we improve the original architecture and achieve position-robust optronic convolutional neural networks (PROPCNNs) with translation invariance through replacing strided convolutional operation and fully connected operation with spectral pooling and global average pooling. PROPCNNs contain less optical components and experiments demonstrate their streamlined architecture is capable of maintaining prediction performance in majority situation of images position variation. This consistent performance provides support for applying our architecture in more complex environment such as loaded on moving vehicles or transported in portable way. • Proposed PROPCNN can maintain performance in position variation situation. • The PROPCNN performs better than OPCNN in both simulations and practical experiments. • The architecture containing less optical components is convenient for alignment. • The PROPCNN reduces the influence of inevitable systematic error in calibration. • Implement the global average pooling and spectral pooling operations in optics. [ABSTRACT FROM AUTHOR]

In this paper, a new algorithm is presented to image ground moving targets in a synthetic aperture radar (SAR) system based on range frequency reversal transform-fractional Fourier transform (RFRT-FrFT). In this algorithm, a range compressed signal is initially transformed into the range frequency domain and then RFRT is proposed to directly compensate the range migration via multiplying the signal in the range frequency domain by its reversed data according to the equal interval sampling of range frequency variable, which can significantly decrease the computational complexity in target envelope migration elimination. Then, FrFT is applied to accomplish the target motion parameter estimation after range migration alignment. Finally, a ground moving target is well focused after motion compensation. The effectiveness of the proposed algorithm is validated by both simulated and real SAR data. [ABSTRACT FROM AUTHOR]

Imaging through scattering media has been a pervasive issue across many fields. Recently, the optronic fully convolutional neural network (OP-FCNN), an opto-electronic deep learning (DL) method using cascaded optical convolutional layers, has shown feasibility in speckle reconstruction and significance for imaging through scattering media. However, achieving higher performance requires cascading deeper optical layers, posing challenges for compact system designs. Here, we propose the optronic Transformer (OPT), a compact opto-electronic DL structure proven effective for imaging through scattering media. The core of OPT utilizes spatial light modulators (SLMs) and optical micro-lens array shift systems to implement the attention mechanism in optics, enabling global information acquisition without cascading extensive optical layers. In contrast to the complex optronic designs of existing OP-FCNN, OPT achieves impressive imaging performance armed with the optronic Transformer with single-head attention, significantly reducing computational complexity and the scales of the optical system. Our OPT's average imaging performance on handwritten digits and letters achieves 0.80, 0.88, 0.82, and 15.29 dB for JI, PCC, SSIM, and PSNR, respectively. The proposed OPT demonstrates superiority and low complexity for imaging through scattering media, presenting a promising opto-electronic solution for designing compact and high-performance DL imaging systems. • The OPT realizes self-attention and residual connections in optical systems. • Reconfigurable, scalable, and compact optical end-to-end architecture. • OPT excels at extracting global features, surpassing optical convolution. • Performs nearly all computational tasks on optical platforms. • OPT achieves superior speckle reconstruction performance. [ABSTRACT FROM AUTHOR]

The effects of complicated motion error on synthetic aperture radar have been analysed. A precise frequency‐division is proposed based on the precise angle‐frequency relation. The proposed method completely compensates the residual aperture‐variant motion and can be integrated into the traditional image formation algorithms directly. Both the simulated data and real airborne data have been applied to validate this method. [ABSTRACT FROM AUTHOR]

Imaging through random media is a pervasive issue in many areas. Recently, deep learning (DL) has shown strong capability in speckle reconstruction, especially in extracting the correlation properties of speckle patterns. However, in practical imaging scenarios, most DL methods need numerous training data to adapt to various imaging condition changes. Here, we develop an S-CNN-based phase conjugation method in imaging through random media. Specifically, we propose the amplitude-phase distortion model to characterize the light scattering process. Then, we further develop a scattering convolutional neural network (S-CNN) based on our model to map speckle patterns to scattered light fields in the changed imaging conditions. Finally, we reconstruct the object through phase conjugation. We quantitatively validated the imaging performance of S-CNN-based phase conjugation imaging, meeting the possible changes of practical imaging conditions, encompassing unseen types of different sparse objects, different diffusers, different detection positions, and different apertures on the imaging path. We found that the Jaccard index (JI), Pearson correlation coefficient (PCC), and structural similarity measure (SSIM) in different physical scenarios were increased by more than 21%, 22%, and 8%, compared with digital CNN. The proposed method provides a new optical imaging system that can be adapted to accommodate various condition changes in practical imaging scenarios and paves the way for an all-optical imaging system in imaging through random media. • Deep learning methods need lots of data to adapt to various imaging conditions. • Physical parameters enable S-CNN to map speckles to scattered light fields. • S-CNN-based phase conjugation adapts to accommodate various imaging conditions. • S-CNN-based phase conjugation has better performance in changed conditions. • S-CNN-based phase conjugation does not need large amounts of training data. [ABSTRACT FROM AUTHOR]