Компьютерная оптика

In this paper, we investigated the dynamics of entanglement of two qubits interacting non-resonantly with the thermal field of a one-mode lossless resonator, taking into account the Ising-type direct interaction between qubits. Based on the exact solution of the quantum Liouville equation, we found the density matrix of the system under consideration. With its help, the reduced qubit-qubit density matrix was calculated and the entanglement criteriоn of the two-qubit system, Pres-Horodeсki parameter, was found. It was shown that for the resonance model and separable initial states of qubits, the direct interaction of qubits leads to a significant increase in the maximum degree of their entanglement. It was also found that for the nonresonant interaction between qubits and field, the increase of the maximum degree of entanglement of qubits is much greater than that due to direct interaction. For the original entangled Bell-type state of the qubits, the direct interaction was found to lead to the vanishing of the effect of the sudden death of entanglement in the case of a resonant qubit-field interaction and, conversely, to an increase in this effect for a nonresonant interaction.

The quality of wavefront conjugation in the case of six-wave interaction in two-dimensional multimode waveguides with Kerr nonlinearity is analyzed under the condition that one of the pump waves excites a zero waveguide mode and the amplitude distribution of the other pump wave at the waveguide end facet varies according to the Gauss law. It is shown that in a waveguide with infinitely conducting walls, the half-width of the modulus of the point spread function of a six-wave radiation converter is completely determined by the transverse size of the waveguide and weakly depends on the width of the Gaussian pump wave. In a parabolic-index profile waveguide, a decrease in the width of the Gaussian pump wave at the waveguide ends commonly leads to a monotonic decrease in the half-width of the point spread function modulus.

We consider two analytical models describing resonant scattering of light by resonant diffraction gratings possessing a horizontal symmetry plane. For the case of oblique incidence, a multiple interference model is formulated, on the basis of which, a new approach for deriving the coupled-mode theory for the considered structures is proposed. Both considered models describe resonances in the reflectance and transmittance spectra of the studied diffraction gratings arising due to the excitation of quasiguided modes and Fabry–Pérot modes. Interaction of these modes leads to the appearance of bound states in the continuum, which are also described by the proposed models.

Ключевые слова

Abstract: We obtain explicit analytic expressions for the Ince-Gaussian (IG) beams for several first indices p = 3, 4, 5, 6. Earlier, explicit expressions have been derived for amplitudes of the IG beams with p = 0, 1, 2 and without regard for the ellipticity parameter. Here, we give expressions for the amplitudes of 24 IG beams written as superpositions of the Laguerre-Gaussian (LG) or Hermite-Gaussian (HG) beams, with the superposition coefficients explicitly depending on the ellipticity parameter. Simultaneously expressing the IG modes both via the LG and HG modes allows easily obtaining the IG modes in the extreme cases when the ellipticity parameter is zero or infinite. Explicit dependence of the obtained expressions for the IG modes on the ellipticity allows the intensity pattern at the beam cross-section to be varied by continuously varying the parameter value. For the first time, intensity distributions are obtained for the IG beams with negative ellipticity parameter.

A combined adaptive optical system designed to correct the wavefront of light radiation distorted by the influence of strong atmospheric turbulence is presented. The system consists of a beam position stabilizer and a fast adaptive optical system operating in real time. Stabilization of the beam position is carried out by two electronically controlled tilt mirrors. The control loop includes two quadrant sensors and an FPGA (field-programmable gate array) that closes the feedback loop. Using information received from Shack-Hartmann wavefront sensor, a bimorph-based adaptive mirror, controlled by another FPGA, is able to compensate for wavefront aberrations for up to 23-th Zernike polynomial in real time. The system was tested under the laboratory turbulence conditions created by a fan heater. The bandwidth of the artificial turbulence in the experiments did not exceed 100 Hz, which corresponds to the average statistical state of the real atmosphere. Results of the correction of the wave front distorted by the artificial turbulence are presented. It is shown that when only a bimorph corrector is used, the value of wavefront tilt angles increases. This presents a certain problem, since a significant part of the turbulence energy falls on the wave front tilts. To address this problem, it is proposed additionally using a system for stabilizing the position of the light beam.

Based on the Richards-Wolf formalism, we investigate tight focusing of a generalized Poincaré beam. Analytical expressions are derived for all components of the electric field strength vector in the focal plane. For superposition of a right-handed circularly polarized plane wave with a left-handed circularly polarized optical vortex with topological charge –1, we obtain expressions for the intensity distribution and the longitudinal component of the spin angular momentum vector at the tight focus. We demonstrate both theoretically and numerically that the initial beam has topological charge –1/2 and, in the center of the focal plane, there is a C-point (a point with circular polarization which makes a C-line along the optical axis) with the singularity index of –1/2 (star), whereas the vector of major axis of the polarization ellipse makes along a certain-radius circle around the optical axis a single-side polarization Möbius stripe of the order –3/2, which has three half-turnovers and one stitching, where two opposite major axis vectors of the polarization ellipse meet.

We study the transformation of the analyzed object by a variety of spatial filters including bandpass, differential, radial filters, various types of vortex filters, and Hilbert filters. Advantages and disadvantages of different filters in terms of clarity and direction-invariance of edge extraction, as well as the energy efficiency are numerically demonstrated. Based on the results obtained, multi-order optical spatial vortex filters with different parameters are developed for simultaneously extracting contours of various object parts. We show that a multi-order filter makes it possible to form a set of images in one plane with a clearly defined contour, object corners and various parts of the contour. Numerical and experimental testing of 4- and 5-channel spatial vortex filters of various types was applied for test objects. A possibility of using the proposed filters to simultaneously highlight the edges of the entire image and various parts of the image in order to extract more features from the analyzed image is shown.

In the scope of a computational experiment, high-contrast gratings (HCG) formed on a silicon-on-insulator (SOI) platform within vertical-cavity surface-emitting lasers (VCSELs) were studied for multispectral laser sources. A simulation model for spectral characteristics calculation is proposed, which includes two heterogeneously integrated parts of the VCSEL: 1) the lower output mirror based on a HCG grating in the silicon layer of the SOI surrounded by air cavities to enhance the contrast of the HCG; 2) the semiconductor VCSEL structure with an air aperture for current and optical confinement. Comparative analysis results of the spectral characteristics of VCSEL-SOI structures for zeroth, first, and second-order modes, which can be excited in the air aperture of the VCSEL, are presented. It is demonstrated that the HCG, acting as one of the cavity mirrors, effectively discriminates the VCSEL higher-order modes. An algorithm for calculating HCG parameters that ensure the maximum reflectivity at a fixed thickness of the silicon layer of the SOI is developed.

Presentation of folded documents is not an uncommon case in modern society. Digitizing such documents by capturing them with a smartphone camera can be tricky since a crease can divide the document contents into separate planes. To unfold the document, one could hold the edges potentially obscuring it in a captured image. While there are many geometrical rectification methods, they were usually developed for arbitrary bends and folds. We consider such algorithms and propose a novel approach Unfolder developed specifically for images of documents with a crease from folding in half. Unfolder is robust to projective distortions of the document image and does not fragment the image in the vicinity of a crease after rectification. A new Folded Document Images dataset was created to investigate the rectification accuracy of folded (2, 3, 4, and 8 folds) documents. The dataset includes 1600 images captured when document placed on a table and when held in hand. The Unfolder algorithm allowed for a recognition error rate of 0.33, which is better than the advanced neural network methods DocTr (0.44) and DewarpNet (0.57). The average runtime for Unfolder was only 0.25 s/image on an iPhone XR.

Ключевые слова

This paper examines a presentation attack when a color photo of a gray copy of a document is presented instead of the original color document during remote user identification. To detect such an attack, we propose an algorithm based on the comparison of chromaticity histograms of presented color images of the document and the ideal template of this type of document. The chromaticity histograms of the original document and the template are expected to be quite identical, while the histograms of the gray copy of the document and the template would be different. The algorithm was tested on images from the open dataset DLC-2021, which contains color images of synthesized identity documents and color images of their gray copies. The precision of the proposed method was 98.99 %, the recall was 84.7 %, that gave 8 times fewer errors than the baseline provided by authors of DLC-2021.

Ключевые слова

The focus of this paper is the study of modern non-blind image deconvolution methods and their application to practical tasks. The aim of the study is to determine the current state-of-the-art in non-blind image deconvolution and to identify the limitations of current approaches, with a focus on practical application details. The paper proposes approaches to examine the influence of various effects on the quality of restoration, the robustness of models to errors in blur kernel estimation, and the violation of the commonly assumed uniform blur model. We developed a benchmark for validating non-blind deconvolution methods, which includes datasets of ground truth images and blur kernels, as well as a test scheme for assessing restoration quality and error robustness. Our experimental results show that those neural network models lacking any pre-optimization, such as quantization or knowledge distillation, fall short of classical methods in several key properties, such as inference speed or the ability to handle different types of blur. Nevertheless, neural network models have made notable progress in their robustness to noise and distortions. Based on the results of the study, we provided recommendations for more effective use of modern non-blind image deconvolution methods. We also developed suggestions for improving the robustness, versatility and performance quality of the models by incorporating additional practices into the training pipeline.

Ключевые слова

Binary neural networks (BNNs) have gained attention due to their computational efficiency. However, training BNNs has proven to be challenging. Existing algorithms either fail to produce stable and high-quality results or are overly complex for practical use. In this paper, we introduce a novel quantizer called UBQ (Uncertainty-based quantizer) for BNNs, which combines the advantages of existing methods, resulting in stable training and high-quality BNNs even with a low number of trainable parameters. We also propose a training method involving gradual network freezing and batch normalization replacement, facilitating a smooth transition from training mode to execution mode for BNNs. To evaluate UBQ, we conducted experiments on the MNIST and CIFAR-10 datasets and compared our method to existing algorithms. The results demonstrate that UBQ outperforms previous methods for smaller networks and achieves comparable results for larger networks.

Ключевые слова

The demand for transmitting video data is increasing annually, necessitating the use of high-quality equipment for reception and processing. The paper presents a neural network recognition system for videos transmitted via a binary symmetrical channel. The presence of digital noise in the data makes it challenging to recognize objects in videos even with advanced neural networks. The proposed system consists of a noise interference detector, a noise purification system based on an adaptive median filter, and a neural network for recognition. The experiment results demonstrate that the proposed system effectively reduces video noise and accurately identifies multiple objects. This versatility makes the system applicable in various fields such as medicine, life safety, physics, and chemistry. The direction of further research may be to improve the model neural network, increasing the database for training or using other noises for modeling.

Existing approaches to barcode detection have a number of disadvantages, including being tied to specific types of barcodes, computational complexity or low detection accuracy. In this paper, we propose YOLO-Barcode – a deep learning model inspired by the You Only Look Once approach that allows to achieve high detection accuracy with real-time performance on mobile devices. The proposed model copes well with a large number of densely spaced barcodes, as well as highly elongated one-dimensional barcodes. YOLO-Barcode not only successfully detects the huge variety of barcode types, but also classifies them. Comparing with the previous universal barcode detector DilatedModel based on semantic segmentation, the YOLO-Barcode is 4 times faster and achieves state-of-the-art accuracy on the ZVZ-real public dataset: 98.6% versus 88.9% by F1-score. The analysis of existing publicly available datasets reveals the absence of many real-life scenarios of mobile barcode reading. To fill this gap, the new “SE-barcode” dataset is presented. The proposed model, used as a baseline, achieves a 92.11% by F1-score on this dataset.

Ключевые слова

Computed tomography (CT) is widely utilized for analyzing internal structures, but the limitations of traditional reconstruction algorithms, which often require a large number of projections, restrict their effectiveness in time-critical tasks or for biological objects studying. Recently Monitored reconstruction approach was proposed for reducing the requirement of dose load. In this paper, there were investigated the advantages of using post-processing neural networks within a monitored reconstruction approach. Three algorithms, namely FBP, FBPConvNet, and LRFR, are evaluated based on their mean count of projections required for the achievement of target reconstruction accuracy. A novel training method specifically designed for neural network algorithms within the Monitored reconstruction framework is proposed. It is shown that the use of the LRFR approach allows one to achieve both a reduction in the number of measured projections and an improvement in the reconstruction accuracy over a certain range of stopping rules. These findings highlight the significant potential of neural networks to be used in the Monitored reconstruction approach.

Ключевые слова

The paper considers the problem of satellite multimodal image registration, in particular, optical and SAR (Synthetic Aperture Radar). Such algorithms are used in object detection, change detection, navigation. The paper considers algorithms for optical-to-SAR image registration in conditions of rough image pre-alignment. It is known that optical and SAR images have an inaccuracy in registration with georeference (up to 100 pixels with a spatial resolution of 10 m/pixel). This paper presents a neural network algorithm for optical-to-SAR image registration based on descriptors calculated for a uniform grid of points. First, algorithm find uniform grid of points for both images. Next, the neural network calculates descriptors for each point and finds descriptor distances between all possible pairs of points between optical and SAR images. Using obtained descriptor distances, a matching is made between the points on the optical and SAR images. The found matches between points are used to calculate the geometric transformation between images using the RANSAC algorithm with a limited (to combinations of translation, rotation and uniform scaling) affine transformation model. The accuracy of the proposed algorithm for optical-to-SAR image registration was investigated with different distortions in rotation and scale.

This paper is dedicated to the proposed techniques of modelling artificial neural networks (NNs) by application of the multigrammatical framework. Multigrammatical representations of feed-forward and recurrent NNs are described. Application of multiset metagrammars to modelling deep learning of NNs of the aforementioned classes is considered. Possible developments of the announced approach are discussed.

The purpose of the article is to create an effective method for low-delay monitoring of the operating state of a drill string and a drill bit without interfering with the proper drilling process. For the drilling process to be continuously controlled, an experimental setup that operates by utilizing the phase-metric method of control was developed. Any movement of the bit causes a change in the electrical characteristics of the probing signal. To obtain a stable signal from a bit immersion depth of up to 250 m, a frequency of probing electrical signals of 166 Hz and an amplitude of up to 500 V were used; the sampling rate of an analog-to-digital converter (ADC) was 10101 Hz. To identify the state of the drill string and the bit based on graphs of time-dependences of changes in the probing signal electrical characteristics, the present authors investigated a number of deep learning methods. Based on the results of the study, a series of capsular neural network methods ( CapsNet ) was chosen. The authors developed modifications of 2D-CapsNet: windowed Fourier transform (WFT) - 2D-CapsNet and frequency slice wavelet transform (FSWT) - 2D-CapsNet. Both of these methods showed a 99% accuracy in determining the transition between two layers of rocks with different properties, which is 2–3% higher than the currently used measurement-while-drilling (MWD) and logging-while-drilling (LWD) rock surveys. Both of these methods unambiguously reveal self-oscillations in the drill string. When determining a fully serviceable bit in the case of self-oscillations, the (FSWT) - 2D-CapsNet method showed an accuracy of 99%.