DSpace :: browse.title.page

Документ

Відкритий доступ

Advanced Method of Land Cover Classification Based on High Spatial Resolution Data and Convolutional Neural Network

(Anhalt University of Applied Sciences, 2022) Shelestov, Andrii; Yailymov, Bohdan; Yailymova, Hanna; Shumilo, Leonid; Lavreniuk, Mykola; Lavreniuk, Alla; Sylantyev, Sergiy; Kussul, Nataliia

Документ

Відкритий доступ

Agriculture land appraisal with use of remote sensing and infrastructure data

(IEEE, 2022) Kussul, Nataliia; Shelestov, Andrii; Yailymova, Hanna; Shumilo, Leonid; Drozd, Sophia

Документ

Відкритий доступ

Automatic Deforestation Detection based on the Deep Learning in Ukraine

(IEEE, 2021) Shumilo, Leonid; Lavreniuk, Mykola; Kussul, Nataliia; Shevchuk, Bella

Документ

Відкритий доступ

Biophysical Impact of Sunflower Crop Rotation on Agricultural Fields

(Sustainability, 2022) Kussul, Nataliia; Deininger, Klaus; Shumilo, Leonid; Lavreniuk, Mykola; Ayalew Ali, Daniel; Nivievskyi, Oleg

Документ

Відкритий доступ

Complex method for land degradation estimation

(IOP Conf. Series: Earth and Environmental Science, 2023-01) Kussul, Nataliia; Shumilo, Leonid; Yailymova, Hanna; Shelestov, Andrii; Krasilnikova, Tetiana

Документ

Відкритий доступ

Crop Yield Forecasting for Major Crops in Ukraine

(IEEE, 2021) Shelestov, Andrii; Shumilo, Leonid; Yailymova, Hanna; Drozd, Sophia

Документ

Відкритий доступ

Generative adversarial network augmentation for solving the training data imbalance problem in crop classification

(2023) Shumilo, Leonid; Okhrimenko, Anton; Kussul, Nataliia; Drozd, Sofiia; Shkalikov, Oleh

Deep learning models offer great potential for advancing land monitoring using satellite data. However, they face challenges due to imbalanced real-world data distributions of land cover and crop types, hindering scalability and transferability. This letter presents a novel data augmentation method employing Generative Adversarial Neural Networks (GANs) with pixel-to-pixel transformation (pix2pix). This approach generates realistic synthetic satellite images with artificial ground truth masks, even for rare crop class distributions. It enables the creation of additional minority class samples, enhancing control over training data balance and outperforming traditional augmentation methods. Implementing this method improved the overall map accuracy (OA) and intersection over union (IoU) by 1.5% and 2.1%, while average crop type classes’ user accuracy (UA) and producer accuracies (PA), as well as IoU, were improved by 11.2%, 6.4% and 10.2%.

Документ

Відкритий доступ

Generative Adversarial Networks for the Satellite Data Super Resolution Based on the Transformers with Attention

(2023) Lavreniuk, Mykola; Shumilo, Leonid; Lavreniuk, Alla

In recent years, free access to high and medium resolution data has become available, providing researchers with the opportunity to work with low resolution satellite images on a global scale. Sentinel-1 and Sentinel-2 are popular sources of information due to their high spectral and spatial resolution. To obtain a final product with a resolution of 10 meters, we have to use bands with a resolution of 10 meters. Other satellite data with lower resolution, such as Landsat-8 and Landsat-9, can improve the results of land monitoring, but their harmonization requires a process known as super-resolution. In this study, we propose a method for improving the resolution of low-resolution images using advanced deep learning techniques called Generative Adversarial Networks (GANs). The state-of-the-art neural networks, namely transformers, with the combination of channel attention and self-attention blocks were employed at the base of the GANs. Our experiments showed that this approach can effectively increase the resolution of Landsat satellite images and could be used for creating high resolution products.

Документ

Відкритий доступ

Geospatial monitoring of sustainable and degraded agricultural land

(2023-07) Yailymova, Hanna; Yailymov, Bohdan; Kussul, Nataliia; Shelestov, Andrii; Shumilo, Leonid

In this study, the assessment of sustainable development goal (SDG) indicator 2.4.1 for Ukraine and Germany is conducted using geospatial and satellite data. The traditional methodology for the SDG indicator 2.4.1 calculation cannot be directly applied to the Ukrainian territory due to the lack of systematic data collection of the essential indicators. Therefore, the authors have developed an integrated approach to estimate land degradation, that uses different schemes for various land cover and crop types at the national scale, utilizing satellite data and employing the WOFOST model for crop growing simulation. The research describes the information sources used for creation crop type classification maps and the necessary data for modeling leaf area index (LAI) based on the WOFOST model. The calculated indicators are determined for each Ukrainian region from 2018 to 2022. Observations in 2022 show a decline in the indicator 2.4.1 across nearly all regions of Ukraine, directly attributed to the military conflicts within the Ukraine. To assess the possibility of applying the developed technology to a large area, the indicator was calculated for a European country (Germany).

Документ

Відкритий доступ

Google Earth Engine Framework for Satellite Data-Driven Wildfire Monitoring in Ukraine

(MDPI, 2023-10) Yailymov, Bohdan; Shelestov, Andrii; Yailymova, Hanna; Shumilo, Leonid

Wildfires cause extensive damage, but their rapid detection and cause assessment remains challenging. Existing methods utilize satellite data to map burned areas and meteorological data to model fire risk, but there are no information technologies to determine fire causes. It is crucially important in Ukraine to assess the losses caused by the military actions. This study proposes an integrated methodology and a novel framework integrating burned area mapping from Sentinel-2 data and fire risk modeling using the Fire Potential Index (FPI) in Google Earth Engine. The methodology enables efficient national-scale burned area detection and automated identification of anthropogenic fires in regions with low fire risk. Implemented over Ukraine, 104.229 ha were mapped as burned during July 2022, with fires inconsistently corresponding to high FPI risk, indicating predominantly anthropogenic causes.

Документ

Відкритий доступ

Ground Based Validation of Copernicus Atmosphere Monitoring Service Data for Kyiv

(IEEE, 2021) Shelestov, Andrii; Yailymova, Hanna; Yailymov, Bohdan; Samoilenko, Oleg; Shumilo, Leonid

Документ

Відкритий доступ

Information Technology for Land Degradation Assessment Based on Remote Sensing

(Anhalt University of Applied Sciences, 2022) Kussul, Nataliia; Shelestov, Andrii; Shumilo, Leonid; Titkov, Dmytro; Yailymova, Hanna

Документ

Відкритий доступ

Is Soil Bonitet an Adequate Indicator for Agricultural Land Appraisal in Ukraine?

(Sustainability, 2021) Shumilo, Leonid; Lavreniuk, Mykola; Skakun, Sergii; Kussul, Nataliia

Документ

Відкритий доступ

Losses Assessment for Winter Crops Based on Satellite Data and Fuzzy Logic

(IEEE, 2021) Bilokonska, Yuliia; Yailymova, Hanna; Yailymov, Bohdan; Shelestov, Andrii; Shumilo, Leonid; Lavreniuk, Mykola

Документ

Відкритий доступ

Monitoring of Fires Caused by War in Ukraine Based on Satellite Data

(IEEE, 2023) Yailymov, Bohdan; Yailymova, Hanna; Shelestov, Andrii; Shumilo, Leonid

The focus of this paper is on fire monitoring studies which utilize a variety of satellite data. The study examines various data sources that are used to automatically detect fires at a national level in Ukraine. Existing fire monitoring systems have low spatial resolution, which makes it difficult to detect fires. Therefore, the authors propose an approach that uses both low- and high-resolution satellite data for fire monitoring. The study found that MODIS, Landsat-8,9 and Sentinel-2 satellite data provide reliable information for fire monitoring in Ukraine. These data sources offer a variety of benefits, including high spatial resolution, frequent revisit times, and wide spectral coverage. In order to understand how favourable weather conditions are for the occurrence of fire, the authors used the Fire Potential Index (FPI). The methodology developed in this study provides a promising approach to monitoring fires and understanding the causes of fires, particularly those caused by the war in Ukraine. The authors implemented the fire detection methodology and FPI index assessment in the Google Earth Engine cloud platform, which allowed for efficient processing and analysis of large volumes of data. In conclusion, the paper highlights the importance of using a combination of low- and high-resolution satellite data for fire monitoring.