Shadow detection is a fundamental and complex problem in the field of computer vision and image processing. Increasing computing power has allowed many deep-learning approaches to be used to solve this problem. In this article, we consider a DSDNet neural network for shadow detection based on texture analysis of a shaded area and a bright area of an urban environment.

The study aims to evaluate the Landsat 8 Thermal Infrared Sensor land surface temperature (LST) products and Visible Infrared Imager Radiometer Suite (VIIRS) LST environmental data record against ground observations in Armenia and Belarus. The data from the Landsat-8 satellite is used due to its high spatial resolution. In contrast, the VIIRS information is obtained due to its high temporal resolution, accessible with a minimal delay after the flyby of satellites. Algorithms to calculate LST were analyzed to find the best performance for both daytime and nighttime data. The evaluation shows that the current VIIRS LST products demonstrate a reasonable accuracy, with a root mean squared error average of 2.77 $\mathbf{K}$ and an average coefficient of determination of 0.92 K.

In this article, we formalize the problem of semiautomatic construction of the contour of area objects from satellite hyperspectral images and present a solution algorithm using PCA and Dijkstra’s algorithm. The contour is considered as the boundary of an object, which can be used for its segmentation and classification. The semiautomatic contour accepts reference points specified by the operator. The formalization of the algorithm is completed.

Over time, the population and anthropogenic activities increase, adversely affecting our atmosphere. In particular, an increase in air temperature leads to a change in the number of various gases that fill the atmosphere, the growth and decrease of which can significantly affect the environment. Nitrogen dioxide (NO₂) is a common air pollutant from natural and human activities, with potential respiratory irritant effects at high levels. To effectively address these issues, there is a need for advanced tools and services that can analyze and visualize NO₂ air pollution data comprehensively. This article presents the development and implementation of a Data Analytics and Visualization Service tailored to Armenia and Belarus’s unique air quality dynamics. The service aims to provide actionable insights for policymakers, researchers, and the public. The studies show that NO₂ excesses in their troposphere for Armenia and Belarus. This increase is correlated with the rise in urban population, causing dramatic adverse effects on the atmosphere.