Computer Optics

643-653

Authors: V.V. Kotlyar, A.G. Nalimov, S.S. Stafeev, A.A. Kovalev

It is theoretically and numerically shown that when tightly focusing an n-th order vector light field that has the central V-point (at which the linear polarization direction is undetermined), the polarization singularity index n, and a "flower"-shaped intensity pattern with 2(n-1) lobes it forms a transverse intensity distribution with 2(n-1) local maxima. At the same time, a vector light field with the polarization singularity index -n, which has the form of a "web" with 2(n+1) cells generates at the sharp focus a transverse intensity distribution with 2(n+1) local maxima. In the focal spot, either 2(n-1) or 2(n+1) V-point polarization singularities with alternating indices +1 or -1 are formed at the intensity zero.

661-666

Authors: E.A. Bibikova, N.D. Kundikova, A.A. Shulginov, N. Al-Wassiti

The spin angular momentum and the extrinsic orbital angular momentum of light are associated with the polarization of light and the light propagation trajectory, respectively. Those momenta are interdependent not only in an inhomogeneous or anisotropic medium but even in free space. This interaction is called the spin-orbit interaction of light. The effects of the spin-orbit interaction of light manifest themselves in a small transverse shift of the beam field longitudinal component from the beam propagation axis in the waist region under the circular polarization sign change. They can be observed both for Gaussian beams and for structured beams. The effects of the spin-orbit interaction of light should be taken into account when nanophotonics devices are created, but the detailed investigation of the effect had not been performed yet due to the low intensity noise image of the beam waist. Precise measurements of the focal waist centerline are needed to determine the transverse shift of the beam field longitudinal component of the asymmetric converging beam's waist under the circular polarization sign change. We propose methods for determining the transverse and longitudinal positions of the beam waist. Computer image processing methods made it possible to obtain the value of the beam waist's transverse position with an accuracy of 0.1 mkm. These methods will allow further testing of the shifts' theoretical predictions, the values of which are the order of 1 mkm. The results obtained can also be used for laser processing of materials by polarized light and precise positioning of the beam's focal spot at a surface.

673-677

Authors: V.S. Pavelyev, K.N. Tukmakov, A.S. Reshetnikov, V.V. Gerasimov, N.D. Osintseva, B.A. Knyazev

Experimental results of the investigation of self-healing properties of terahertz Bessel beams with orbital angular momentum (OAM) with topological charges of l=3 and l=4 in free space after passing through a dispersive medium are presented.

685-691

Authors: S.S. Stafeev, V.D. Zaicev

It is shown theoretically and numerically that circularly and linearly polarized incident beams produce at the tight focus identical circularly symmetric distributions of an on-axis energy flux. It is also shown that the on-axis energy fluxes from radially and azimuthally polarized optical vortices with unit topological charge are equal to each other. An optical vortex with azimuthal polarization is found to generate the minimum focal spot measured for the intensity (all other parameters being equal). Slightly larger (by a fraction of a percent) is the spot size calculated for the energy flux for the circularly and linearly polarized light. The spot size in terms of intensity is of importance in light-matter interaction, whereas the spot size in terms of energy flux affects the resolution in optical microscopy.

702-712

Authors: D.V. Tropin, A.M. Ershov, D.P. Nikolaev, V.V. Arlazarov

The demand for on-device document recognition systems increases in conjunction with the emergence of more strict privacy and security requirements. In such systems, there is no data transfer from the end device to a third-party information processing servers. The response time is vital to the user experience of on-device document recognition. Combined with the unavailability of discrete GPUs, powerful CPUs, or a large RAM capacity on consumer-grade end devices such as smartphones, the time limitations put significant constraints on the computational complexity of the applied algorithms for on-device execution. In this work, we consider document location in an image without prior knowledge of the document content or its internal structure. In accordance with the published works, at least 5 systems offer solutions for on-device document location. All these systems use a location method which can be considered Hough-based. The precision of such systems seems to be lower than that of the state-of-the-art solutions which were not designed to account for the limited computational resources. We propose an advanced Hough-based method. In contrast with other approaches, it accounts for the geometric invariants of the central projection model and combines both edge and color features for document boundary detection. The proposed method allowed for the second best result for SmartDoc dataset in terms of precision, surpassed by U-net like neural network. When evaluated on a more challenging MIDV-500 dataset, the proposed algorithm guaranteed the best precision compared to published methods. Our method retained the applicability to on-device computations.

721-727

Authors: A.D. Fida, A.V. Gaidel, N.S. Demin, N.Yu. Ilyasova, E.A. Zamytskiy

We discuss approaches to combining multimodal multidimensional images, namely, three-dimensional optical coherence tomography (OCT) data and two-dimensional color images of the fundus. Registration of these two modalities can help to adjust the position of the obtained OCT images on the retina. Some existing approaches to matching fundus images are based on finding key points that are considered invariant to affine transformations and are common to the two images. However, errors in the identification of such points can lead to registration errors. There are also methods for iterative adjustment of conversion parameters, but they are based on some manual settings. In this paper, we propose a method based on a full or partial search of possible combinations of the OCT image transformation to find the best approximation of the true transformation. The best approximation is determined using a measure of comparison of preprocessed image pixels. Further, the obtained transformations are compared with the available true transformations to assess the quality of the algorithm. The structure of the work includes: pre-processing of OCT and fundus images with the extraction of blood vessels, random search or grid search over possible transformation parameters (shift, rotation and scaling), and evaluation of the quality of the algorithm.

736-748

Authors: A.S. Konushin, B.V. Faizov, V.I. Shakhuro

Traffic sign recognition is a well-researched problem in computer vision. However, the state of the art methods works only for frequent sign classes, which are well represented in training datasets. We consider the task of rare traffic sign detection and classification. We aim to solve that problem by using synthetic training data. Such training data is obtained by embedding synthetic images of signs in the real photos. We propose three methods for making synthetic signs consistent with a scene in appearance. These methods are based on modern generative adversarial network (GAN) architectures. Our proposed methods allow realistic embedding of rare traffic sign classes that are absent in the training set. We adapt a variational autoencoder for sampling plausible locations of new traffic signs in images. We demonstrate that using a mixture of our synthetic data with real data improves the accuracy of both classifier and detector.

756-766

Authors: Yu.V. Vizilter, O.V. Vygolov, S.Yu. Zheltov

We introduce attribute and relational representations of mosaic image models with directed relationships between regions. Attribute representations of asymmetric relational models based on stacking, ranking and integral descriptions are considered. We propose some morphological shape similarity measures based on relational models. We show that using the same oriented relational model, various morphological operators can be constructed, in particular, of Serra- or Pyt’ev type. Some constructive methods for the design of such morphological operators in an attribute and relational domains are proposed. From this consideration we also extract a new morophlogical scheme for two-stage mutual adaptive image-and-shape joint filtering: at the first step, the shape is simplified (projected) with regard to the image to be projected, and at the second step, the image is simplified (projected) with regard to the simplified (projected) shape.

773-778

Authors: A.S. Mokeev, V.M. Yamshchikov

We discuss features of the calculation of a Fraunhofer integral by traditional quadrature numerical integration methods and a special collocation Levin method when calculating the diffraction of a plane electromagnetic wave by a rectangular aperture. For the quadrature numerical integration methods, a criterion for the assessment of the integration step is derived depending on the screen size and required calculation accuracy. Advantages of the use of the special collocation Levin method in comparison with the traditional quadrature numerical integration methods are shown.