Optical Memory and Neural Networks最新文献

We have obtained an expression for the amplitude and intensity of the field along the caustic line, which is formed during autofocusing when the generalized lens is illuminated. Two types of the incident beam amplitude are considered: the product of a power and generalized exponential function; fractional-rational amplitude without singularities. With different ratios between the parameters, the dependence of intensity on distance will have a diverse form – increasing, decreasing, with a maximum, with a minimum. At the same time, the edge values can also be different: zero, positive number, infinity.

Transverse dynamics of a broad-area semiconductor surface-emitting laser with a vertical resonator (VCSEL) with еxternal optical injection was investigated above threshold. It is obtained that two types of instabilities can develop in system: Turing and Hopf instabilities. Turing instability in turn can be subdivided into plane-wave instability (PWI) and modulation instability (MI). We found some sets of system parameters where the steady-state curve of the homogeneous solution can be bistable or monostable. Modulation instability can lead to the development of spatial patterns, which was obtained.

In this paper, studies have been carried out on the formation of complex structured Laguerre-Gauss vector beams. For the first time, a digital method of intensity moments was used, which allows one to determine the mode composition using a single intensity distribution in each polarization component in the focal plane, either of a spherical lens, or in the plane of double focus of a cylindrical lens.

The paper investigates the sensitivity of interferograms formed using the structured reference beams. The parameters of the reference beam are selected to improve the visualization of aberrations in the interferograms. A study carried out on the use of reference beams with cylindrical wavefronts in the interferograms formation to improve the aberrations recognition using a convolutional neural network. The applying of a cylindrical reference beam instead of a plane one in the interference method for recognition of wave aberrations based on neural networks with Xception architecture makes it possible to reduce the mean absolute error by more than 30%. In this work, for each type of interferogram, the model was trained for 80 epochs, which took about 1.8 hours using GeForce RTX 2070 graphics card. However, after completing this training once, we obtain a model that allows us to make forecasts in 0.055 s for every new interferogram of the same type.

The family of Poincare beams has three characteristics including two real-valued angular characteristics, which descibe a concrete polarisation state on the Poincare sphere and a third integer characteristic l determining the beam singularity order. It was theoretically and numerically shown that at l = 2, an energy backflow is generated near the optical axis. It was revealed that at certain Poincare beams characteristics, the energy flow revolves around the optical axis because of spin-orbital conversion. It was also demonstrated that a radial optical Hall phenomenon took place in the compact focus of Poincare beams.

We investigate the efficiency of convolutional neural networks (CNNs) application for recognition of two-mode optical vortex (OV) beams superpositions. Unlike standard multiplexing, we associate information channels not with individual modes, but with pairs of modes with a given index difference which raises security of information transmission. At the first stage, we performed studies with a model dataset using standard image augmentation techniques for training CNNs (translation and rotation). Further, we use experimentally generated by phase spatial light modulator (SLM) intensity patterns for training the proposed neural networks. The achieved test accuracy of the CNNs trained on the experimentally generated dataset is 0.84. This value is comparable with the test accuracy for the modeling training dataset.

In this paper, we present the main results of the numerical calculation of the bound states in the continuum (BICs) application in the refractive index (RI) sensing. We consider two types of one-dimensional rectangular photonic structure at subwavelength regime and study the Fano resonance shift for off-Γ BIC. The results obtained will make it possible to create photonic structures for biosensing, for which a high-Q resonance is observed at nonzero incidence angles.

This paper investigates the accuracy of convolutional neural network recognition of Laguerre and Hermite-Gauss optical modes with geometric distortions in the form of affine transformations. The influence of training data sampling on the accuracy is analyzed. The ability of a convolutional neural network to recognize Laguerre and Hermite-Gaussian optical modes with geometric distortions caused by environmental distortions and described by affine transformations was also examined.

Two imaging systems are considered in this paper: a system with a lens containing an obstacle in the front focal area of the lens, and a system with an image in the front focal area of the lens with a phase mask and a circular aperture diaphragm. A rotating Laguerre–Gaussian beam propagates through the first imaging system, and structure of the beam transforms after propagation through an obstacle. The simulations of the system were performed using Fresnel transformations for the field propagation before and after the lens. The simulation results showed how much the beam is distorted at the output of the imaging system depending on the size of the obstacle and its distance from the propagation axis. Similarly, a rotating Laguerre–Gaussian beam with a superimposed “triangle” shadow image propagates through the second system. The simulations were performed using the Fresnel transform and overlaying a limiting aperture in the lens plane in order to investigate the influence of the aperture size on the sharpness of the generated image.

This paper investigates structured Laguerre-Gaussian (sLG) beams with two control parameters under an astigmatic transformation. The presented beams are a structurally stable superposition of the standard Laguerre Gaussian mode and the hybrid Hermite–Laguerre–Gaussian mode. By changing the phase parameter of such beam, one can obtain wave solutions that are stable under simple astigmatism. We also found sharp spikes and dips in orbital angular momentum (OAM) in astigmatic sLG beams in the region where orbital angular momentum vanishes. The height and depth of bursts and dips significantly exceed the maximum and minimum of OAM values in ordinary sLG beams for a given phase parameter.