Computer Optics

705-713

Authors: L.L. Doskolovich, A.A. Mingazov, D.A. Bykov, E.A. Bezus

A problem of calculating a refractive surface that forms a required irradiance distribution in the far field in the case of a plane illuminating beam is considered. We show that this problem can be formulated as a mass transportation problem. The specific form of the cost function for this problem is obtained. It is shown that with a certain choice of coordinates, the cost function becomes quadratic. The resulting mass transportation problem also describes a problem of calculating a mirror, which can be considered as a special case of the problem of calculating a refractive surface.

723-734

Authors: A.V. Volyar, M.V. Bretsko, Ya.E. Akimova, Yu.A. Egorov, V.V. Milyukov

Transformations of the vortex beams structure subjected to sectorial perturbation were theoretically and experimentally studied. The analysis was based on computing (measuring) the vortex spectrum that enables us to find the orbital angular momentum (OAM) and Shannon entropy (informational entropy). We have revealed that, in the general case, the number of vortices caused by an external perturbation is not related to the topological charge. For arbitrary perturbation, the topological charge remains equal to the initial topological charge of the unperturbed vortex beam. Growth of the vortex number induced by perturbations is associated with the optical uncertainty principle between the sectorial angle and the OAM. The computer simulation has shown that OAM does not depend on the number of vortices induced by perturbations. Moreover, two maxima are formed both in the positive and negative regions of the vortex spectrum. As a result, the OAM does not practically change in a wide range of perturbation angles from 0 to 90 °. However, at large perturbation angles, when the energy is almost equally redistributed between the vortex modes with opposite signs of the topological charge, the OAM rapidly decreases. At the same time, the Shannon entropy monotonically increases with growing perturbation angle. This is due to the fact that the entropy depends only on the number of vortex states caused by external perturbations.

741-746

Authors: B. Sokolenko, N. Shostka, O. Karakchieva, A.V. Volyar, D. Poletayev

In present paper we propose easy way to implement method of interfering vortices with opposite topological charge for the real time determination of the thickness and information about the surface of studied samples with the resolution up to 7 nanometers. The determination of the characteristics of the medial cross-section of submicron-objects becomes possible due to phase sensitivity of interfering singular beams to the slightest changes in the optical path difference between them. The dependence of rotational angle of resulting interference pattern in case of different sample thickness for two singular beams superposition is considered in detail.

756-764

Authors: S.N. Khonina, K.N. Tukmakov, S.A. Degtyarev, A.S. Reshetnikov, V.S. Pavelyev, B.A. Knyazev, Yu.Yu. Choporova

A silicon subwavelengh terahertz axicon has been designed, fabricated, and investigated by methods of numerical and optical experiments. The research has been performed on a free-electron laser workstation NOVOFEL (Budker Institute of Nuclear Physics of SB RAS, Novosibirsk). Diffractive and polarization features of realized element have been investigated.

773-779

Authors: S.P. Murzin, M.V. Blokhin

Experimental studies of shaping a CO2 laser beam with a reflective diffractive optical element have been performed. To increase the aperture of the initial beam, we used a collimator containing a system of two plane-convex spherical ZnSe lenses. For the focal line formed with the diffractive optical element in combination with the collimator, in addition to a 1.3-fold increase in the length, a decrease in the maximum beam power density was found to occur in the laser spot. It was demonstrated that under the laser action it is possible to generate in a two-phase steel sample regions of full hardening, selective hardening, and annealing, alongside the initial structure. The formation of such structures is due to the distribution pattern of temperature fields and a difference in the cooling rate across the volume of the heat affected zone.

789-795

Authors: V.S. Vasilev, R.V. Skidanov, S.V. Ganchevskaya

We consider the possibility of using generalized lenses in imaging optical systems. We propose using two generalized lenses in a system similar to a Fourier-correlator to obtain a point-to-point mapping, where the second lens is set so that a plane wave is also formed at the output of the system when the incident wave is plane. We show that for generalized lenses with a small deviation of their degree from 2, such a system forms a fairly high-quality image. We use a standard radial photographic chart to test the system, which allows us to determine its key optical characteristics.

803-809

Authors: М.P. Kalaev, А.М. Telegin, K.E. Voronov, Jiang Lixiang, Jiao Jilong

The paper describes a DF-OPTICS device that the present authors designed for the experimental study of changes in the optical properties of a glass whose surface is exposed to high-speed flows of micron-sized dust particles. The device allows the scattering indicatrix and the spectral transmittance to be automatically measured at each point of the sample with a 0.5-mm increment. Advantages of the developed device include small dimensions and the ability to work in vacuum, allowing it to be used in an accelerator chamber to simulate outer space factors. Experimental results for the K-8 glass put in a microparticle accelerator and bombarded by an aluminum powder PAP-1 with a characteristic size of 1-3 µm and speeds of 2-8 km / s are presented. The device makes it possible to measure the change of the spectral transmittance of transparent materials in the UV and RGB regions with an accuracy of 0.005%, which exceeds the sensitivity of some known spectrophotometers.

818-824

Authors: V.V. Arlazarov, K. Bulatov, T. Chernov, V.L. Arlazarov

A lot of research has been devoted to identity documents analysis and recognition on mobile devices. However, no publicly available datasets designed for this particular problem currently exist. There are a few datasets which are useful for associated subtasks but in order to facilitate a more comprehensive scientific and technical approach to identity document recognition more specialized datasets are required. In this paper we present a Mobile Identity Document Video dataset (MIDV-500) consisting of 500 video clips for 50 different identity document types with ground truth which allows to perform research in a wide scope of document analysis problems. The paper presents characteristics of the dataset and evaluation results for existing methods of face detection, text line recognition, and document fields data extraction. Since an important feature of identity documents is their sensitiveness as they contain personal data, all source document images used in MIDV-500 are either in public domain or distributed under public copyright licenses. The main goal of this paper is to present a dataset. However, in addition and as a baseline, we present evaluation results for existing methods for face detection, text line recognition, and document data extraction, using the presented dataset.

833-845

Authors: K.S. Sarin, I.A. Hodashinsky

Handwritten signature verification is an important research area in the field of person authentication and biometric identification. There are two known methods for handwriting signature verification: if it is possible to digitize the speed of pen movement, then verification is said to be on-line or dynamic; otherwise, when only an image of handwriting is available, verification is said to be off-line or static. It is proved that when using dynamic verification, a greater accuracy is achieved than when using static verification. In the present work, the amplitudes, frequencies, and phases of the harmonics extracted from the signature signals of the X and Y coordinates of the pen movement using a discrete Fourier transform are used as characteristics of the signature. All signals are pre-processed in advance, including the elimination of gaps, the elimination of the angle of inclination, the normalization of position and scaling. A fuzzy classifier is proposed as a signature verification tool based on the features obtained. The work examines the effectiveness of this tool in the ensemble, as well as using a procedure for feature selection. To build an ensemble of classifiers, a well-known bagging method is used, and the feature selection is based on the determination of mutual information between a feature and a class of an object. Experiments on signature verification on the SVC2004 data set with the construction of a fuzzy classifier and ensembles of three, five, seven and nine fuzzy classifiers were conducted. Experiments were carried out both with the use of the feature selection procedure and without selection. The efficiency of the classifiers constructed is compared with each other and with known analogues: decision trees, support vector machines, discriminant analysis and k-nearest neighbors.

857-868

Authors: N.I. Chervyakov, P.A. Lyakhov, N.N. Nagornov, M.V. Valueva, G.V. Valuev

Modern convolutional neural networks architectures are very resource intensive which limits the possibilities for their wide practical application. We propose a convolutional neural network architecture in which the neural network is divided into hardware and software parts to increase performance and reduce the cost of implementation resources. We also propose to use the residue number system in the hardware part to implement the convolutional layer of the neural network for resource costs reducing. A numerical method for quantizing the filters coefficients of a convolutional network layer is proposed to minimize the influence of quantization noise on the calculation result in the residue number system and determine the bit-width of the filters coefficients. This method is based on scaling the coefficients by a fixed number of bits and rounding up and down. The operations used make it possible to reduce resources in hardware implementation due to the simplifying of their execution. All calculations in the convolutional layer are performed on numbers in a fixed-point format. Software simulations using Matlab 2017b showed that convolutional neural network with a minimum number of layers can be quickly and successfully trained. Hardware implementation using the field-programmable gate array Kintex7 xc7k70tfbg484-2 showed that the use of residue number system in the convolutional layer of the neural network reduces the hardware costs by 32.6% compared with the traditional approach based on the two’s complement representation. The research results can be applied to create effective video surveillance systems, for recognizing handwriting, individuals, objects and terrain.

886-900

Authors: Yu.V. Vizilter, V.S. Gorbatsevich, S.Y. Zheltov

A general approach to a structure-functional analysis and synthesis (SFAS) of deep neural networks (CNN). The new approach allows to define regularly: from which structure-functional elements (SFE) CNNs can be constructed; what are required mathematical properties of an SFE; which combinations of SFEs are valid; what are the possible ways of development and training of deep networks for analysis and recognition of an irregular, heterogeneous data or a data with a complex structure (such as irregular arrays, data of various shapes of various origin, trees, skeletons, graph structures, 2D, 3D, and ND point clouds, triangulated surfaces, analytical data descriptions, etc.) The required set of SFE was defined. Techniques were proposed that solve the problem of structure-functional analysis and synthesis of a CNN using SFEs and rules for their combination.