Search

The paper considers the problems of macroparameter control in a linear dynamic system. In the authors’ previous paper, such problems in controlling nonlinear systems were investigated. Variations of the considered functionals and necessary optimality conditions were obtained. This paper focuses on the trajectory beam control in the linear case. In this case, the problem of beam control can be reduced to the problem of the individual trajectory control. The developed approach may be of interest in the study of charged particle dynamics in accelerating and focusing structures, as well as in the construction of the velocity field in automatic image processing.

This paper deals with problems of program control of a trajectory beam. We investigate formulations that arise when considering problems of charged particle beams control, as well as, for example, in image processing. In applied problems, it is often necessary to investigate the problem of the gravity center control of some set or changing the density distribution of particles according to some given one. The functionals proposed in the paper can be effectively used for the velocity field constructing in various image processing, in particular, medical diagnostic images. In this paper, the problem of constructing a velocity field is considered as a control and optimization problem, and, unlike the previous works of the authors, macroparameters characterizing the objects under study are used in optimization. In the article, variations of the studied functionals are obtained and the necessary optimality conditions are given.

Artificial intelligence has become widespread in image processing tasks. At the same time, the number of vulnerabilities is increasing in systems implementing these artificial intelligence technologies (the attack surface is increasing). The main threats to information security can be implemented by introducing malicious perturbations into the input data, regardless of their type. To detect such attacks, approaches and methods have been developed based, in particular, on the use of an auto-encoder or the analysis of layers of the target neural network. The disadvantage of existing methods, which significantly reduce the scope of their application, is binding to the dataset or model architecture. This paper discusses the issues of expanding the scope (increasing scalability) of methods for detecting L0-optimized perturbations introduced by unconventional pixel attacks. An approach to detecting these attacks using statistical analysis of input data, regardless of the model and dataset, is proposed. It is assumed that the pixels of the perturbation embedded in the image, as a result of the L0-optimized attack, will be considered both local and global outliers. Outlier detection is performed using statistical metrics such as deviation from nearest neighbors and Mahalanobis distance. The evaluation of each pixel (anomaly score) is performed as a product of the specified metrics. A threshold clipping algorithm is used to detect an attack. When a pixel is detected for which the received score exceeds a certain threshold, the image is recognized as distorted. The approach was tested on the CIFAR-10 and MNIST datasets. The developed method has demonstrated high accuracy in detecting attacks. On the CIFAR-10 dataset, the accuracy of detecting onepixel attack (accuracy) was 94.3 %, and when detecting a Jacobian based Saliency Map Attack (JSMA) — 98.3 %. The proposed approach is also applicable in the detection of modified pixels. The proposed approach is applicable for detecting one-pixel attacks and JSMA, but can potentially be used for any L0-optimized distortions. The approach is applicable for color and grayscale images regardless of the dataset. The proposed approach is potentially universal for the architecture of a neural network, since it uses only input data to detect attacks. The approach can be used to detect images modified by unconventional adversarial attacks in the training sample before the model is formed.

In this paper a new method for contour detection in grayscale images is proposed. The pro-posed method is based on the use of an image weight model, which allows one to estimate its pix-els from the point of view of their significance for perception. In this case, the most significant pixels are those that contain characteristic features of the image, including brightness differences at the boundaries of the regions. To assess the significance of pixels, we propose a procedure for analyzing the contribution of the corresponding wavelet coefficients at different scale levels to the total energy of the image. The described method of contour detection involves building an image weight model, determining the directions of linear segments along the edges on the weight image, analyzing the significance of pixels and linking significant pixels. The advantage of the method is the high operation speed (the corresponding loop detector works on average four times faster than the Canny edge detector). In addition, the paper describes a detector of significant image areas, which is also based on the weight model. The proposed approach can be used in various systems of information processing and control based on methods and tools of computer vision, including control and navigation systems of unmanned vehicles, remote sensing of the Earth, systems for pavement defect detection, biometric systems, etc.

This paper suggests an approach to the semantic image analysis for application in computer vision systems. The aim of the work is to develop a method for automatically construction of a semantic model, that formalizes the spatial relationships between objects in the image and research thereof. A distinctive feature of this model is the detection of salient objects, due to which the construction algorithm analyzes significantly less relations between objects, which can greatly reduce the image processing time and the amount of resources spent for processing. Attention is paid to the selection of a neural network algorithm for object detection in an image, as a preliminary stage of model construction. Experiments were conducted on test datasets provided by Visual Genome database, developed by researchers from Stanford University to evaluate object detection algorithms, image captioning models, and other relevant image analysis tasks. When assessing the performance of the model, the accuracy of spatial relations recognition was evaluated. Further, the experiments on resulting model interpretation were conducted, namely image annotation, i.e. generating a textual description of the image content. The experimental results were compared with similar results obtained by means of the algorithm based on neural networks algorithm on the same dataset by other researchers, as well as by the author of this paper earlier. Up to 60 % improvement in image captioning quality (according to the METEOR metric) compared with neural network methods has been shown. In addition, the use of this model allows partial cleansing and normalization of data for training neural network architectures, which are widely used in image analysis among others. The prospects of using this technique in situational monitoring are considered. The disadvantages of this approach are some simplifications in the construction of the model, which will be taken into account in the further development of the model.

Subject and Purpose. In this paper, an algorithm is developed to improve the spectrum image resolution in the atomic emission spectrum registration using multi-element detectors and taking multiple exposures with image shifting by a fractional part of a pixel. Methods and Methodology. Mathematical modeling techniques are employed in the coordinate representation dictated by the prior information. The modeling results are checked by experiment. Results. The importance of the resolution value of spectral image registration systems and its impact on the quality of the obtained results have been highlighted. The spectral informativeness of the regularly (matrix- or linearly) structured multi-element detectors has been evaluated against the irregular-structure sensors, such as photographic emulsions. As it has been found that the process of pixel size reduction is demanding, an alternative approach through subpixel image shifting has been proposed. Unlike the case of the corresponding instruments implemented as a rule in the frequency domain, here the relevant theoretical problem reduces to the over- determined system of linear equations. The proposed solution algorithm employs the least squares method having regard to the prior data specificity posed by the isolation of the spectral lines and a smooth flow of the background between them. The digital simulation results and the real experimental evidence from the atomic emission spectral analysis have been presented, illustrating the algorithm operation. The experimental research was carried out for one-dimensional spectrum images registered with linear charge-coupling photodetectors. The so gained resolution was twice the spatial resolution of the photodetector. Conclusions. It has been demonstrated that controlled image shifting not only improves the photometry accuracy but also significantly enhances the detector resolution. The general comparisons of the theory with the experiment have proved the feasibility of bringing the proposed methodology into atomic emission analysis.

A problem of restoration of a blurred image obtained by a horizontally rotating camera is considered in this paper. The mathematical model of this problem is the convolution equation on a cyclic group. In the previous papers of the authors, the case of a non-singular equation was considered. The general case that admits the degeneracy of the convolution equation is considered in this paper. An algorithm is developed based on which a fast program for recovering blurred images is written. The complexity of the algorithm presented in this paper for the degenerate convolution equation is the same as for the non-singular case. The analysis of calculation errors affecting the image quality is given. The influence of the errors of the algorithm's initial data for a degenerate equation is not higher than for a non-singular case.

Copyright of Radio Electronics, Computer Science, Control is the property of Zaporizhzhia National Technical University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Two new methods of adaptive median filtering of impulse noise in images are proposed in the paper. The first method is based on the joint application of iterative processing and transformation of the result of median filtering using the Lorentzian function. The second method uses alternative masks of median filter, calculated using the Euclidean metric. This approach has made it possible to reduce the size of the processed area without the loss in quality for low-intensity noise. The experimental part of the article shows the results of comparison of the performance of the proposed methods with the known methods. We used three different images distorted by impulse noise with pixel distortion probabilities ranging from 1 % up to 99 %. The numerical evaluation of the quality of image denoising based on the peak signal to noise ratio (PSNR) and the structural similarity (SSIM) has shown that the proposed method shows a better result of processing in all the cases considered, as compared with the known approaches. The results obtained in the paper may find wide practical applications when processing satellite and medical imagery, geophysical data, and in other areas of digital image processing.

The problem of determining the velocity field is investigated. This problem is considered by many authors in various formulations. The most well-known statement of the problem is proposed with the use of a concept of optical flow of constant distribution density function (brightness of images) along trajectories of the system under consideration. In addition, besides the common grey value constancy assumption, also, gradient constancy, as well as the constancy of the Hessian and the Laplacian are considered. In this statement functionals of quality are constructed, that also require the smoothness of the considered velocity field. The minimization of the constructed functional usually reduces to solving the Euler-Lagrange equations by numerical methods. In this paper a new formulation of the problem is proposed. The density along the trajectories is assumed to vary. The velocity field is defined as a function depending on the vector of unknown parameters. In this paper an optimization approach to constructing the velocity field is proposed, which is based on the study of the integral functional on trajectories ensembles. The variation of integral functional is represented in an analytical form, which makes it possible to use gradient methods to find the required parameters. The proposed approach can be used in the analysis of various images, in particular, of radionuclide images.

In the process of development of computer technologies, the number of areas of their application naturally grows and, along with it, the complexity of the tasks to be solved, which entails the need for new research. Similar tasks include digital filtering of images in the field of medical technologies and active-pulse television measuring systems. There are many methods and algorithms of digital filtering designed to solve the problem of improving the quality; algorithms that can improve the quality of images while reducing computational costs are widely used. High demands, which are made due to the constant growth in the size of the generated images, as well as the requirement for modern television systems, is real-time operation. When solving practical problems, it is required to use different filter aperture sizes, which provide an increase in quality and preservation of image details. The solution of these problems was the reason for the emergence of adaptive filters that are able to change the parameters in the process of processing the received data, while not spending additional time on processing with an increase in the size of the aperture. The paper presents the principles of constructing adaptive image processing filters, which, by obtaining an input parameter indicating the required dimension of a multi-element aperture, are able to implement the construction of the required aperture. The Laplacian “Truncated Pyramid” filter and the “double pyramid” Laplacian were modified. A feature of these filters is the oddness of the multi-element aperture, so the coefficient used to build the mask is always set to odd. When using these filters, it is possible to use two coefficients that are responsible for increasing the filtration efficiency, since, in their original form, the Laplacian filters have a sum of coefficients equal to zero. The experiment shows a comparison with high-dimensional filters that work when using classical two-dimensional convolution. The next stage of the presented research will be the application of parallel computing techniques, which will increase the speed of the developed filters.

This paper is devoted to steganalysis algorithm determining presence a secret message embedded into digital image which is stored in compressed or uncompressed form. The algorithm is based on the analysis of pairs of colors and uses the method of modifying the least significant bit. [ABSTRACT FROM AUTHOR]

The paper proposes an approach to the segmentation of solid breast lesions and their peripheral areas in ultrasound images. It is noted that identifying the outermost breast lesion structures is an important step for the further lesion classification, directly affecting the final classification of its type. The main feature of the proposed approach is that its implementation takes into account peculiarities of pixel brightness variations in the original image, without using speckle noise filters. The method was tested on a set of ultrasound images of morphologically verified 42 benign and 49 malignant breast lesions marked by a radiologist. The segmentation results were compared with the results of manual marking performed by the radiologist. The average errors in the segmentation of benign and malignant lesion were 5 pixels – for the lesion area and 7 pixels – for the peripheral area, which is insignificant, taking into account the error of manual marking performed by radiologist (3.9 and 4.7 pixels, respectively). The average intersection-over-union (IoU) metrics were 0.82 and 0.80, respectively. The presented results indicate the possibility of using the developed technology in a combination with the system of lesion differentiation.

Detailed automated analysis of crop images is critical to the development of smart agriculture and can significantly improve the quantity and quality of agricultural products. A hyperspectral camera potentially allows to extract more information about the observed object than a conventional one, so its use can help in solving problems that are difficult to solve with conventional methods. Often, predictive models that solve such problems require a large dataset for training. However, sufficiently large datasets of hyperspectral images of agricultural plants are not currently publicly available. Therefore, we present a new dataset of hyperspectral images of plants in this paper. This dataset can be accessed via URL https://pypi.org/project/HSI-Dataset-API/. It contains 385 hyperspectral images with a spatial resolution of 512 by 512 pixels and spectral resolution of 237 spectral bands. The images were captured in the summer of 2021 in Samara and Novocherkassk (Russia) using Offner based Imaging Hyperspectrometer of our own production. The article demonstrates the work of some basic approaches to the analysis of hyperspectral images using the dataset and states problems for further solving.

The paper proposes a method for detecting and removing impulse noise in images, which determines the similarity between pixels by distance and the difference in brightness values in the local detector window. An impulse noise model is considered, where distorted pixels take random values and also randomly appear in the image. Pixels identified as pulses are recovered with an adaptive median filter. The impulses are determined in the detector window, whose size is calculated in the Euclidean metric and increases with increasing noise intensity in the image. In the experimental part, we discuss visual differences between familiar methods and the one proposed herein on three images for three different impulse noise intensities. In the approximation of image fragments, it is seen that the proposed method copes with the task in the best way, which is also confirmed by numerical estimates of the quality of image reconstruction from impulse noise based on the peak signal-to-noise ratio and the structural similarity index. The proposed method can be used when solving problems of cleaning images under conditions of distorting impulses and for eliminating distortions caused by adverse weather effects, such as raindrops and snow.

Many different filters can be used to reduce multiplicative speckle noise on radar images. Most of these filters have some parameters whose values influence the result of filtering. Finding optimal values of such parameters may be a non-trivial task. In this paper, a formal automated method for finding optimal parameters of speckle noise reduction filters is proposed. Using a specially designed test image, optimal parameters for the most commonly used filters were found using several image quality assessment metrics, including the Structural Similarity Index (SSIM) and Gradient Magnitude Similarity Deviation (GMSD). The use of filters with optimal parameters allows processing (detection, segmentation, etc.) of radar images with minimal influence of speckle noise.

Continuous improvement of methods for visual information registration, processing and storage leads to the need of improving technical characteristics of digital image processing systems. The paper proposes new high-performance digital filter architectures for image processing by the Winograd method with calculations performed in a residue number system with special-type moduli. To assess the performance and hardware costs of the proposed architectures, hardware simulation is carried out using a field-programmable gate array in a computer-aided design envi-ronment Xilinx Vivado 2018.3 for the target device Artix-7 xc7a200tffg1156-3. The results of hardware simulation show that the proposed filter architectures have 1.13 – 5.42 times higher performance, but require more hardware costs compared to the known methods. The results of this study can be used in the design of complex systems for image processing and analysis for their performance to be increased.

One of main issues in face identification is the lack of training data of specific type (bad quality image, varying scale or illumination, children/old people faces, etc.). As a result, the recogni-tion accuracy may be low for input images which are not similar to the majority of images in the dataset used to train the feature extractor. In this paper, we propose that this issue is dealt with by the automatic detection of such out-of-distribution data based on the addition of a preliminary stage of their automatic rejection using a special convolutional network trained using a set of rare data collected using various transformations. To increase the computational efficiency, the decision about the presence of a rare image is made on the basis of the same face descriptor that is used in the classifier. Experimental research confirmed the accuracy improvement of the proposed approach for several datasets of faces and modern neural network descriptors.

In this paper, we investigate the effect of squeeze-and-excitation blocks on improving the classification quality of osteoarthritis using convolutional neural networks of the ResNet and DenseNet families. We show that the use of these blocks improves the quality of osteoarthritis classification according to the Kellgren-Lawrence scale by 1–3 % without a significant modifi-cation of the model structure. We also demonstrate that combining the 0 and 1 classes of the Kellgren-Lawrence scale into one class allows one to increase the accuracy of osteoarthritis grading by 12.74 %, without losing significant information about the disease. The best final ac-curacy attained was 84.66 % when using an ensemble of three convolutional networks with the DenseNet-121 architecture using squeeze-and-excitation blocks, which significantly exceeds the performance of the existing state-of-the-art. The obtained results can be used both for a prelimi-nary automatic diagnosis and as an auxiliary tool.

Videoendoscopic and tomographic research are the two leading medical imaging solutions for detecting, classifying and characterizing a wide array of pathologies and conditions. However, source information from these types of research is very different, making it hard to cross-correlate them. The paper proposes a novel algorithm for combining results of based on 3D surface reconstruction methods. This approach allows to align separate parts of two input 3D surfaces: surface obtained by applying bundle adjustment-based 3D surface reconstruction algorithm to the endoscopic video sequence, and surface reconstructed directly from separate tomographic cross-section slice projections with regular density. Proposed alignment method is based on using local feature extractor and descriptor algorithms by applying them to the source surface normal maps. This alignment allows both surfaces to be equalized and normalized relative to each other. Results show that such an adjustment allows to reduce noise, correct reconstruction artifacts and errors, increase representative quality of the resulting model and establish correctness of the reconstruction for hyperparameter tuning.

The problem of a displacement field calculation for an image sequence based on a discrete model is being solved. Algorithms for velocity field (displacement field) construction are in demand in various image processing tasks. These methods are used in motion detection, object movement tracking, analysis of complex images, movement correction of medical diagnostic images in nuclear medicine, radiology, etc. An optimization approach to the displacement field construction based on a discrete model is developed in the paper. The approach explores the possibility of taking into account the brightness change along the trajectories of the system. A linear model is considered. Directed optimization methods based on the analytical representation of the functional gradient are constructed to search for unknown parameters. The algorithm for displacement field construction with image partitioning into regions (neighborhoods) is proposed. This algorithm can be used to process a variety of image sequences. The results of the algorithm operation on test radionuclide images are presented.

The paper considers a problem of classifying Sentinel-2 multispectral satellite images for environmental monitoring of the Baikal Natural Territory (BNT). The specificity of the BNT required the creation of a new set of 12 classes, which takes into account current problems. The set was formed in such a way that the areas corresponding to these classes completely covered the BNT. A training dataset was formed using a web interface based on Sentinel-2 satellite images. The classification of satellite images was carried out using Random Forest algorithms and the ResNet50 neural network. The accuracy of the calculations showed that the classification results can be used to solve actual problems of the Baikal natural territory, in particular, to analyze changes in the forestland, assess the impact of climate change on the landscape, analyze the dynamics of development activities, create farmland inventory, etc.

The relevance of the tasks of detecting and recognizing objects in images and their sequences has only increased over the years. Over the past few decades, a huge number of approaches and methods for detecting both anomalies, that is, image areas whose characteristics differ from the predicted ones, and objects of interest, about the properties of which there is a priori information, up to the library of standards, have been proposed. In this work, an attempt is made to systematically analyze trends in the development of approaches and detection methods, reasons behind these developments, as well as metrics designed to assess the quality and reliability of object detection. Detection techniques based on mathematical models of images are considered. At the same time, special attention is paid to the approaches based on models of random fields and likelihood ratios. The development of convolutional neural networks intended for solving the recognition problems is analyzed, including a number of pre-trained architectures that provide high efficiency in solving this problem. Rather than using mathematical models, such architectures are trained using libraries of real images. Among the characteristics of the detection quality assessment, probabilities of errors of the first and second kind, precision and recall of detection, intersection by union, and interpolated average precision are considered. The paper also presents typical tests that are used to compare various neural network algorithms.

Currently, the determination of the psychoemotional state of the observed, including students and pupils, is an urgent task that has social significance. The pace of technology development and the increased interest of foreign and domestic specialists show that the automation of the determination of psychoemotional reactions is an actual and popular area of research. The article provides an overview of various literature sources on this topic, analyzes the existing types of emotional artificial intelligence that allow a computer to recognize and interpret human emotions and respond to them. The camera reads the person's state, and the neural network processes the data to determine the emotion. In addition, recent research papers published during 2019 and 2020 are included to present the latest advances in this field.

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.

This paper proposes a new switching median filter for suppressing multi-pixel impulse noise in X-ray images. A multi-pixel impulse is understood as a set of several neighboring pixels, the intensity of each significantly exceeds background intensity. Multi-pixel noise can occur, for example, due to the blooming effect, the reason being the limited value of pixel saturation capacity. This article defines the thresholds for the intensity increment relative to the eight immediate neighbors, above which the current pixel is processed by the median filter. The dependence of these thresholds on the number of pixels in an impulse is presented. The proposed algorithm is based on the median filtering process, which consists of several iterations. In this case, the filter has the smallest possible size, which minimizes image distortion during processing. In particular, to exclude a single-pixel impulse, pixel processing is turned on when intensity surge exceeds 3.5 with the grayscale value ranging from 0 to 1. At the same time, to exclude nine-pixel impulses, three iterations are required with the following thresholds: the first iteration with a threshold 2.0; the second iteration also with a threshold 2.0 and the third iteration with a threshold 3.5. The algorithm proposed was tested on real X-ray images corrupted by multi-pixel impulse noise. The algorithm is not only simple, but also reliable and suitable for real-time implementation and application. The efficiency of the technique is shown in comparison with other known filtering methods with respect to the degree of noise suppression. The main result of the testing is that only the proposed method allows excluding multi-pixel noise. Other advantage of the algorithm is its weak effect on the level of Gaussian noise leading to the absence of image blurring (or preserving image details) during processing.

The paper investigates the applicability of the convolutional neural network "U-Net" to a problem of segmentation of aircraft images. The neural network image segmentation method is based on the "Carvana" implementation with the "U-Net" architecture. For orientation recognition, a neural network built in the Keras open neural network library based on the pretrained VGG16 neural network is used. The approach considered allows the image segmentation to be conducted. The results of the experiments have shown the possibility of a fairly accurate selection of the object of interest. The resulting binary masks make it possible to visually classify the aircraft in the image.

Shadow detection is a fundamental challenge in the field of computer vision. It requires the network to understand the global semantics and local details of the image. All existing methods depend on the aggregation of the features of a multi-stage pre-trained convolution neural network but in comparison to high-level capabilities, low-level capabilities provide less to the detection performance. Using low-level features not only increases the complex difficulty of the network but also reduces the time efficiency. In this article, we propose a new shadow detector, which only uses high-level features and explores the complementary information between adjacent feature layers. Experiments show that the technique in this paper can accurately detect shadows and perform well compared with the most advanced methods. The detailed experiments performed in three public shadow detection datasets SUB, UCF, and ISTD demonstrate that the suggested method is efficient and stable.

The paper presents results of the analysis of long-term changes in the vegetation cover of abandoned agricultural lands in the forest-steppe zone of the Central Chernozem Region using time series of the Normalized Difference Vegetation Index (NDVI), which are measured using MOD13Q1 data. The vegetation index dynamics linked with the proportion of forest communities formed on the abandoned agricultural lands is investigated. The index values for the period of mid-August are the most informative for analyzing the share of forest communities growing on the abandoned agricultural land. Abandoned agricultural lands with coniferous forests have a higher correlation with NDVI than fallows with deciduous species. In the period 2000-2018, for all types of abandoned arable lands, the presence of a positive statistically significant trend component of the vegetation index long-term series is established. Using a slope angle coefficient of the NDVI trend line, a spatio-temporal analysis of the rate of formation of forest stands in the forest-steppe fallows at the beginning of the XXI century was carried out. Features of this process are studied.

Many recent studies have focused on developing image reconstruction algorithms in optical systems based on flat optics. These studies demonstrate the feasibility of applying a combination of flat optics and the reconstruction algorithms in real vision systems. However, additional causes of quality loss have been encountered in the development of such systems. This study investigates the influence on the reconstructed image quality of such factors as limitations of mass production technology for diffractive optics, lossy video stream compression artifacts, and specificities of a neural network approach to image reconstruction. The paper offers an end-to-end deep learning-based image reconstruction framework to compensate for the additional factors of quality losing. It provides the image reconstruction quality sufficient for applied vision systems.

The paper discusses different approaches to image and video processing aiming to solve the problems of detecting, tracking and estimating the parameters of moving objects. The developed algorithms for solving these problems are described in relation to the field of transport analytics. When developing the algorithms, attention was given to solving the problems on an embedded platform of video surveillance cameras, which imposes restrictions on the computational complexity. The first (basic) algorithm for moving object detection and parameter estimation is based on processing two associated areas of an image. This algorithm includes a computationally efficient adaptive procedure for evaluating and updating the background component of an image. The procedure is based on the physics of the process of movement of the object of interest through a processing zone. The second algorithm performs object tracking based on an optical flow method initialized by feature points. The third algorithm is based on object segment tracking and is computationally efficient for the implementation on an embedded platform of intelligent cameras. Results of experimental studies of the proposed algorithms are presented, as well as a comparison with some well-known algorithms. It is shown that tracking algorithms can improve the accuracy of moving object parameter estimation. Tracking also reduces the number of classification errors compared to the basic approach to object detection and parameter estimation.

The paper deals with the τ-generalized (by Wright) Gauss hypergeometric function Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed. We formulate and prove the lemma of composition relations for Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed and its adjacent functions. Our proof is based on using the representation of Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed function in the series form as well as employing some properties of the classic gamma function. The proved formulas generalize known expressions for classic Gauss hypergeometric function Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed, a specific case of the considered τ -generalized function Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed when τ = 1 . [ABSTRACT FROM AUTHOR]

This paper considers the structure of a set of systems of continuous solutions in a number of cases depending on the hypotheses for the matrices A, B, number q and their properties. Using the methods of the theory of differential and difference equations, we define new conditions for the existence of continuous solutions of these systems of equations. Specifically, we develop the method of their construction and examine their properties. In theorems 1 and 3 we obtain the results under the conditions ai1, i = 1,…, n, q > 1, (t≤0), 0 i < 1, i = 1,…,m, q > 1, (t ≥ 0), whilst in theorems 5, 6 we obtain the research results under Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed. [ABSTRACT FROM AUTHOR]

The key objective of this paper is to obtain differential formulas for q-integral representation of (τ, β)-generalized hypergeometric function. To this end, we consider new (τ, β)-generalized hypergeometric functions Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed. Using the integral property of q-beta function for the function Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed, we obtain the q-integral representation. Employing the body of the theory of fractional differentiation, we also investigate a series of q-differential relations for these new generalized hypergeometric functions. The results obtained allow widely applying Due to image rights restrictions, multiple line equation(s) cannot be graphically displayed functions to solve problems of mathematical physics, differential and integral equations, the theory of probability, mathematical statistics and others. [ABSTRACT FROM AUTHOR]

The paper considers a problem of extracting user preferences based on their photo gallery. We propose a novel approach based on image captioning, i.e., automatic generation of textual descriptions of photos, and their classification. Known image captioning methods based on convolutional and recurrent (Long short-term memory) neural networks are analyzed. We train several models that combine the visual features of a photograph and the outputs of an Long short-term memory block by using Google's Conceptual Captions dataset. We examine application of natural language processing algorithms to transform obtained textual annotations into user preferences. Experimental studies are carried out using Microsoft COCO Captions, Flickr8k and a specially collected dataset reflecting the user’s interests. It is demonstrated that the best quality of preference prediction is achieved using keyword search methods and text summarization from Watson API, which are 8 % more accurate compared to traditional latent Dirichlet allocation. Moreover, descriptions generated by trained neural models are classified 1 – 7 % more accurately when compared to known image captioning models.

The paper is devoted to the development of an effective semantic segmentation algorithm for automation of airport infrastructure labelling in RGB satellite images. This task is addressed using algorithms based on deep convolutional artificial neural networks, as they have proven themselves in a wide range of tasks, including the terrestrial imagery segmentation, where they show consistently high results. A new dataset was labelled for this particular task and a comparative analysis of different architectures and backbones was carried out. A conditional random field model (CRF) was used for postprocessing and accounting of contextual information and neighborhood of objects of different classes in order to eliminate outliers. Features of the solutions applied at all preparatory stages of the algorithm were described, including data preparation, neural network training and post-processing of the training results.

In this paper, we investigate the urgent problem associated with bone age determination using hand X-rays. A combined algorithm for the recognition of radiographs is proposed, which uses simultaneous two neural network models, based on Xception and DenseNet169. The method allows to generalize the knowledge of different medical experts and increases the accuracy of bone age prediction in general.

In recent years, there has been an increasing interest in the classification of edible vegetable oils, examining authenticity and in detecting possible adulteration of high quality, expensive extra virgin olive oils with low-cost edible oils. Classical methods such as gas chromatography, liquid chromatography, Fourier transform infrared and nuclear magnetic resonance, mass spectrometry, and Raman spectroscopy have been widely applied to examine the authenticity of edible oils. De-spite of their high sensitivity and accuracy, these methods are significantly expensive for daily life testing, especially in resource-poor regions. Furthermore, they are time-consuming as samples have to be analyzed in dedicated laboratories. In this paper, we propose a compact, low-cost, port-able smartphone-based spectrometer for testing edible oil authenticity. Using simple laboratory op-tical components and a smartphone, we developed a compact spectrometer which can function in the wavelength range of 400–700 nm with the spectrum/pixel resolution of 0.334 nm / pixel. The images captured by the smartphone were converted into intensity distribution plots versus wave-length. As a proof of concept, the smartphone based spectrometer was utilized to measure the variations in fluorescent intensity of the mixed oils of expensive extra virgin olive oil and low-cost rice oil with different percentages. The results obtained the spectrometer were in good agreement with that from a laboratory spectrometer, thus, confirmed its adequate sensitivity and accuracy. Due to the cost effectiveness, the adequate sensitivity, and the portability, the smartphone based spectrometer can be applied in numerous applications such as in-field testing, lifestyle monitoring, and home diagnostics.

This work is a continuation of the authors’ previous publications, in which the restoration of images obtained with a horizontally rotating camera was considered. In this paper, a mathematical model is constructed for reconstructing blurred images obtained with a camera rotating in the horizontal plane and having its optical axis tilted at an angle to the horizon. The method of image restoration involves constructing a strip of spherical panorama from the original images. Results of numerical experiments that confirm the good quality of the presented image recovery method and the high performance of the developed algorithm are presented.

Road traffic analysis is an important task in many applications and it can be used in video surveillance systems to prevent many undesirable events. In this paper, we propose a new method based on integral optical flow to analyze cars movement in video and detect flow extreme situations in real-world videos. Firstly, integral optical flow is calculated for video sequences based on optical flow, thus random background motion is eliminated; secondly, pixel-level motion maps which describe cars movement from different perspectives are created based on integral optical flow; thirdly, region-level indicators are defined and calculated; finally, threshold segmentation is used to identify different cars movements. We also define and calculate several parameters of moving car flow including direction, speed, density, and intensity without detecting and counting cars. Experimental results show that our method can identify cars directional movement, cars divergence and cars accumulation effectively.

Detection of objects of interest is a crucial step in the automatic analysis of the medical X-ray images. However, medical X-rays are often characterized by the low contrast as well as great variability in range of colours, which makes it more difficult to be analysed by the common methods based on the regions homogeneity principles. In our paper, we present an alternative approach to the contours detection problem that does not require the homogeneity criteria to be satisfied. Our method is based on the identification of edge fragments and elimination of discontinuities between them. Moreover, we describe a numeric criterion for quality evaluation of contours detection. The obtained results can used for diagnosis of abnormalities and diseases, and also as an intermediate step for more sophisticated methods of image analysis.

The paper gives an overview of the algorithms developed to obtain reference lines and angles on X-ray images. These geometrical characteristics are used in the medical analysis of human joints. We propose the algorithm’s modifications based on the analysis of numerous X-ray images. These modifications allowed obtaining a great increase in calculation speed and the improvement of final results quality given by the corresponding application. They also lead to a significant reduction of manual tuning of the program, arising only in the rare cases when the properties of given images differ significantly from the mean ones.

The paper presents a new method for measuring the vortex spectrum (amplitude squares and initial phases) as well as the orbital angular momentum without cuts and splices of the wave front of a vortex array. Such an extension of the measurement capabilities of the combined beam properties is achieved by recording the intensity moments of the light beam as a whole without destroying its internal structure. The correlation coefficient between the initial intensity distribution and the intensity distribution taking into account the experimentally obtained vortex spectrum reaches 91%, which indicates the reliability of the presented method.