The article focuses on the development of a web portal for monitoring and forecasting atmospheric air quality in the Khabarovsk Territory. The study analyzes existing solutions in the field of environmental monitoring, identifying their key shortcomings, such as the lack of real-time data, limited functionality, and outdated interfaces. The authors propose a modern solution based on the Python/Django and PostgreSQL technology stack, which enables the collection, processing, and visualization of air quality sensor data. Special attention is given to the implementation of harmful gas concentration forecasting using a recurrent neural network, as well as the creation of an intuitive user interface with an interactive map based on OpenStreetMap. The article provides a detailed description of the system architecture, including the backend, database, and frontend implementation, along with the methods used to ensure performance and security. The result of this work is a functional web portal that provides up-to-date information on atmospheric air conditions, forecast data, and user-friendly visualization tools. The developed solution demonstrates high efficiency and can be scaled for use in other regions.

Keywords: environmental monitoring, air quality, web portal, forecasting, Django, Python, PostgreSQL, neural networks, OpenStreetMap

The article describes the experience in the field of practice-oriented and career-oriented activities aimed at modernizing modern engineering education, taking into account an interdisciplinary approach. An example of training an engineer of a new formation based on school engineering classes is given. Attention is paid to project activities that can involve students in both creative and production projects. The issues of "career guidance filtering" necessary for selecting the necessary professional qualities of students through a cycle of mini introductory and training cases are highlighted. The emphasis is placed on the contact and information access of parents and schoolchildren to the university's research sector.

Keywords: management of engineering education, project activities, engineering classes, priority 2030, career guidance activities

This paper is devoted to the construction of a visual-inertial odometry system for an unmanned vehicle using both binocular cameras and inertial sensors as an information source, which would be able to simultaneously determine the vehicle's own position and the relative position of other road users. To ensure accurate and continuous localization, it is proposed to use an inertial navigation system and two types of image keypoints. Deep learning models are used to accurately and reliably track keypoints. To achieve efficient and reliable matching of objects between two frames, a multi-level data association mechanism is proposed that takes into account possible errors of various system components. The experimental results demonstrate the feasibility and application potential of the proposed system.

Keywords: multi-object visual-inertial odometry, localization, data association, tracking of 3D dynamic objects

In this paper, methods for estimating one's own position from a video image are considered. A robust two-stage algorithm for reconstructing the scene structure from its observed video images is proposed. In the proposed algorithm, at the feature extraction and matching stage, a random sample based on the neighborhood graph cuts is used to select the most probable matching feature pairs. At the nonlinear optimization stage, an improved optimization algorithm with an adaptive attenuation coefficient and dynamic adjustment of the trust region is used. Compared with the classical Levenberg-Marquard (LM) algorithm, global and local convergence can be better balanced. To simplify the system's decisions, the Schur complement method is used at the group tuning stage, which allows for a significant reduction in the amount of computation. The experiments confirmed the operability and effectiveness of the proposed algorithm.

Keywords: 3D reconstruction，graph-cut, Structure-from-Motion (SfM)，RANSAC，Bundle Adjustment optimization，Levenberg-Marquardt algorithm，Robust feature matching

This paper is devoted to the construction of a robust visual-inertial odometry system for an unmanned vehicle using binocular cameras and inertial sensors as information sources.The system is based on a modified structure of the VINS-FUSION system. Two types of feature points and matching methods are used to better balance the quantity and quality of tracking points. To filter out incorrect matches of key points, it is proposed to use several different methods. Semantic and geometric information are combined to quickly remove dynamic objects. Keypoints of static objects are used to complement the tracking points. A multi-layer optimization mechanism is proposed to fully utilize all point matchings and improve the accuracy of motion estimation. The experimental results demonstrate the effectiveness of the system.

Keywords: robust visual-inertial odometry, localization, road scene, multi-level optimization mechanism

The purpose of the article: to determine the possibility of using file hash analysis using artificial neural networks to detect exploits in files. Research method: the search for exploits in files is carried out based on the analysis of Windows registry file hashes obtained by two hashing algorithms SHA-256 and SHA-512, using three types of artificial neural networks (direct propagation, recurrent, convolutional). The obtained result: the use of artificial neural networks in file hash analysis allows us to identify exploits or malicious records in files; the performance (accuracy) of artificial neural networks of direct propagation and with recurrent architecture are comparable to each other and are much more productive than convolutional neural networks; the longer the length of the file hash, the more reliably it is to identify an exploit in the file.

Keywords: malware, exploit, neural networks, hashing, modeling

The paper examines the case of IntraService incident management system implementation in an organization operating in the digital infrastructure segment. The study focuses on the assessment of changes that occurred in the functioning of the support service based on quantitative and qualitative indicators. The method of comparative analysis of operational parameters before and after the launch of the system is used, accompanied by expert interpretation of internal processes.

Keywords: implementation, system, incident, support, automation, platform, organization, infrastructure, process, integration

This paper provides a survey of metrics used to assess the quality of images generated by generative models. Specialized metrics are required to objectively evaluate image quality. A comparative analysis showed that a combination of different metrics is necessary for a comprehensive evaluation of generation quality. Perceptual metrics are effective for assessing image quality from the perspective of machine systems, while metrics evaluating structure and details are useful for analyzing human perception. Text-based metrics allow for the assessment of image-text alignment but cannot replace metrics focused on visual or structural evaluation. The results of this study will be beneficial for specialists in machine learning and computer vision, as well as contribute to the improvement of generative algorithms and the expansion of diffusion model applications.

Keywords: deep learning, metric, generative model, image quality, image

The article considers the assessment of the suitability of solar radiation data from ERA5 atmospheric reanalysis for forecasting problems in the northern territories. The experimental site of the Mukhrino station (Khanty-Mansiysk Autonomous Okrug), equipped with an autonomous power supply system, was chosen as the object of analysis. A statistical analysis of the annual array of global horizontal insolation data obtained using the PVGIS platform has been carried out. Seasonal and diurnal features of changes in insolation are considered, distribution profiles are constructed, and emissions are estimated using the interquartile range method. It is established that the data are characterized by high variability and the presence of a large number of zero values due to polar nights and weather conditions. The identified features must be taken into account when building short-term forecasting models. The conclusion is made about the acceptable quality of ERA5 data for use in forecasting energy generation and consumption in heating systems.

Keywords: ERA5, solar radiation, horizontal insolation, the Far North, statistical analysis, forecasting, emissions analysis, renewable energy sources, energy supply to remote areas, time series, intelligent generation management

The article describes the features of using a two-layer membrane with the use of injection control fittings in the installation of underground waterproofing. The circumstances preventing the mass application of this technology have been identified, the main part of which is related to the increase in the cost of work at the initial stage. However, the use of the technology is justified because it allows you to localize the location and period of leakage, has increased maintainability and durability.

Keywords: waterproofing, modern waterproofing technologies, double-layer membrane, injection control fittings

The results of studies of the strength properties of plaster solutions with vermiculite on a mixed binder using a multifunctional additive D-5 are presented. A slight acceleration of the setting time of solutions with a multifunctional additive D-5 was found, while the bending and compressive strengths of the solutions did not change significantly after 2 hours and 28 days. Replacing part of the expanded vermiculite fraction 0 – 0.315 mm with volcanic pumice fraction 0 – 0.315 mm in the construction mortar allows to significantly increase their strength characteristics and reduce the cost without a noticeable increase in density. Expressions of the dependence of the compressive strength of the mortars on the calculated total porosity are obtained.

Keywords: construction gypsum, Portland cement, volcanic pumice, expanded vermiculite, D-5 multifunctional additive, plaster composites, compressive and bending strength

The article presents the results of laboratory testing of a precast lintel made of D600 aerated concrete blocks fastened together using assembly adhesive for laying aerated concrete blocks and three reinforcing bars, two bars in the lower zone with a diameter of 20 mm A400 and one bar in the upper zone with a diameter of 16 mm A400. When testing the beam lintel, 3 loading schemes were assembled, each of which was a beam hinged on concrete supports. The obtained data were analyzed. Based on the test results, destructive loads, the nature of destruction were determined and the resulting deformations were assessed. The nature of destruction is similar to that of reinforced concrete beam structures with insufficient transverse reinforcement; destruction in all tested prototypes occurs in the support zone along inclined sections. The bearing capacity of aerated concrete lintels made of piece material is sufficient for use in construction practice.

Keywords: testing, bearing capacity, reliability, deformability, gas blocks, lintels, destruction

This paper explores the content-based filtering approach in modern recommender systems, focusing on its key principles, implementation methods, and evaluation metrics. The study highlights the advantages of content-based systems in scenarios that require deep object analysis and user preference modeling, especially when there is a lack of data for collaborative filtering.

Keywords: сontent - oriented filtering, recommendation systems, feature extraction, similarity metrics, personalization

This article highlights the use of tethered balloons in the construction of high-rise buildings, where the use of cranes or helicopters is not always advisable due to the technological features of the structures being built, or for economic reasons. The design scheme of a balloon system capable of mounting high-rise structures, in particular industrial chimneys, is described. Examples of the installation of high-rise pipe structures are given. The basis of theoretical approaches is given to determine the characteristics of the shell as the main component of cargo balloon carriers. A method for determining the volume of a balloon is given, as an important element in the balloon shell design.

Keywords: balloon crane, construction of high-rise facilities, high-rise installation, balloon system, equation of equilibrium of a balloon crane.

The article presents an analysis of modern methods of image generation: variational autoencoders (VAE), generative adversarial networks (GAN) and diffusion models. The main attention is paid to a comparative analysis of their performance, generation quality and computational requirements. The Frechet Inception Distance (FID) metric is used to assess the image quality. Diffusion models showed the best results (FID 20.8), outperforming VAE (FID 59.75) and GAN (FID 38.9), but require significant resources. VAEs are stable, but generate blurry images. GANs provide high quality, but suffer from training instability and mode collapse. Diffusion models, due to step-by-step noise decoding, combine detail and structure, which makes them the most promising. Also considered are methods of image-to-image generation used for image modification. The results of the study are useful for specialists in the field of machine learning and computer vision, contributing to the improvement of algorithms and expansion of the areas of application of generative models.

Keywords: deepfake, deep learning, artificial intelligence, GAN, VAE, diffusion model