Статті КІТС (ДМетІ)

Permanent URI for this collectionhttp://crust.ust.edu.ua/handle/123456789/14595

Browse

Now showing 1 - 13 of 13

Aircraft Detection in Aerial Imagery Based on YOLO Architectures
(CEUR-WS Team, Aachen, Germany, 2025) Kashtan, Vita Yu.; Radionov, Yevhen; Hnatushenko, Volodymyr V.
ENG: The study is devoted to determining the most efficient YOLO-based architecture for the task of aircraft detection in high-resolution aerial imagery. A comparative analysis was conducted across YOLO models v8 through v11 under three experimental conditions: using pre-trained (raw) models, fine-tuning the models on a domain-specific dataset, and fine-tuning models to a dataset enhanced through a proposed image preprocessing method. The evaluation considered both accuracy and inference performance metrics. The proposed methodology reduced the false negative rate from 19.5% to 3.2% at a confidence threshold of 0.75, underscoring its effectiveness in enhancing target visibility under challenging imaging conditions such as low contrast or background clutter.
Aircraft Detection with Deep Neural Networks and Contour-Based Methods
(National University "Zaporizhzhia Polytechnic", Zaporizhzhia, 2024) Radionov, Y. D.; Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Kazymyrenko, O. V.
ENG: Context. Aircraft detection is an essential task in the military, as fast and accurate aircraft identification allows for timely response to potential threats, effective airspace control, and national security. The use of deep neural networks improves the accuracy of aircraft recognition, which is essential for modern defense and airspace monitoring needs. Objective. The work aims to improve the accuracy of aircraft recognition in high-resolution optical satellite imagery by using deep neural networks and a method of sequential boundary traversal to detect object contours. Method. A method for improving the accuracy of aircraft detection on high-resolution satellite images is proposed. The first stage involves collecting data from the HRPlanesv2 dataset containing high-precision satellite images with aircraft annotations. The second stage consists of preprocessing the images using a sequential boundary detection method to detect object contours. In the third stage, training data is created by integrating the obtained contours with the original HRPlanesv2 images. In the fourth stage, the YOLOv8m object detection model is trained separately on the original HRPlanesv2 dataset and the dataset with the applied preprocessing, which allows the evaluation of the impact of additional processed features on the model performance. Results. Software that implements the proposed method was developed. Testing was conducted on the primary data before preprocessing and the data after its application. The results confirmed the superiority of the proposed method over classical approaches, providing higher aircraft recognition accuracy. The mAP50 index reached 0.994, and the mAP50-95 index reached 0.864, 1% and 4.8% higher than the standard approach. Conclusions. The experiments confirm the effectiveness of the proposed method of aircraft detection using deep neural networks and the process of sequential boundary traversal to detect object contours. The results indicate this approach’s high accuracy and efficiency, which allows us to recommend it for use in research related to aircraft recognition in high-resolution images. Further research could focus on improving image preprocessing methods and developing object recognition technologies in machine learning.
Automated Building Damage Detection on Digital Imagery Using Machine Learning
(Dnipro University of Technology, Ukraine, 2023) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.
ENG: Purpose. To develop an automated method based on machine learning for accurate detection of features of a damaged building on digital imagery. Methodology. This article presents an approach that employs a combination of unsupervised machine learning techniques, specifically Principal Component Analysis (PCA), K-means clustering, and Density-Based Spatial Clustering of Applications with Noise (DBSCAN), to identify building damage resulting from military conflicts. The PCA method is utilized to identify principal vectors representing the directions of maximum variance in the data. Subsequently, the K-means method is applied to cluster the feature vector space, with the predefined number of clusters reflecting the number of principal vectors. Each cluster represents a group of similar blocks of image differences, which helps to identify significant features associated with fractures. Finally, the DBSCAN method is employed to identify areas where points with similar characteristics are located. Subsequently, a binary fracture mask is generated, with pixels exceeding the threshold being identified as fractures. Findings. The introduced methodology attains an accuracy rate of 98.13 %, surpassing the performance of conventional methods such as DBSCAN, PCA, and K-means. Furthermore, the method exhibits a recall of 82.38 %, signifying its ability to effectively detect a substantial proportion of positive examples. Precision of 58.54 % underscores the methodology’s capability to minimize false positives. The F1 Score of 70.90 % demonstrates a well-balanced performance between precision and recall. Originality. DBSCAN, PCA and K-means methods have been further developed in the context of automated detection of building destruction in aerospace images. This allows us to significantly increase the accuracy and efficiency of monitoring territories, including those affected by the consequences of military aggression. Practical value. The results obtained can be used to improve automated monitoring systems for urban development and can also serve as the basis for the development of effective strategies for the restoration and reconstruction of damaged infrastructure.
Computer Modeling of Territory Flooding in the Event of an Emergency at Seredniodniprovska Hydroelectric Power Plant
(Dnipro University of Technology, Ukraine, 2022) Ivanov, D. V.; Hnatushenko, Volodymyr V.; Kashtan, Vita Yu.; Garkusha, I. M.
ENG: Purpose. Computer modeling of territory flooding in the event of an emergency at Seredniodniprovska Hydroelectric Power Plant (HPP). Methodology. The computer model of possible territory flooding at Seredniodniprovska HPP is developed using simulation modeling methods and geometric and hydrological approaches and considers initial boundary conditions of the water-engineering system. Calculations of the wave break height and the half-divided cross-sectional area of the river bed were made and a three-dimensional model of the territory flooding was built using the Python language and ArcGIS Desktop software. Findings. The data for each creation of the hydraulic node, namely the depth and width of the flooded territory, were calculated. This allowed analyzing the macro level considering the triangulation model of the surface. The wave break parameters and flaps (intersections) were taken into account in case of a dam break at a hydroelectric power plant or a rise in the water level. A mathematical model, and a 3D model were developed, and a forecast of the flood zone due to an emergency was made using satellite survey data. Originality. The mathematical method received further development for calculating flood territories in the event of an emergency at Seredniodniprovska Hydroelectric Power Plant, taking into account the parameters of the breakthrough wave and the calculation of cross-sections for the cases when a hydroelectric dam breaks or the water level rises; the method uses one-dimensional and two-dimensional systems of Saint-Venant equations, and geometric and hydrological approaches. A three-dimensional model of the territory flooding is developed to predict possible consequences. Practical value. The obtained results can be used to model the flooding of the territory located near dangerous hydro-technical objects, such as dams, dikes as well as to forecast flooded territories during the construction of drainage and protective structures.
Deep Learning-Based Segmentation of Multi-Temporal Satellite Imagery for Flood Detection
(CEUR-WS Team, Aachen, Germany, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.
ENG: A deep learning-based pixel-based flood zone segmentation approach is proposed using multi-temporal satellite images and topographic and hydrological information. It is proposed to combine heterogeneous data (satellite images before and after the flood, digital elevation model, and hydrographic characteristics) into a single input tensor, allowing the neural network to consider the area's spatial and temporal dynamics and morphometric features. The architecture of the model ensures the preservation of the spatial detail of the flooded area through skip-connection mechanisms, which contributes to the correct identification of flood boundaries. Comparative analysis with FCNN, DeepLabv3, and BASNet confirmed the superiority of the proposed approach (F1-score 82%, Dice 82% for the category 'flooded areas'), which indicates its effectiveness for accurately detecting flooded areas.
Entropy-Cryptographic Approach for Transmission of Satellite Data in Telecommunication Networks
(Khmelhytskyi National University, Khmelhytskyi, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.
ENG: Amid rapid advances in satellite observations and increasing risks associated with high-resolution multispectral data downloads, stronger satellite image encryption techniques are becoming increasingly important. To address these challenges, the paper proposes a new cryptographic methodology based on entropy and nonlinear methods. This approach offers high resistance against brute-force attacks, generates a high-entropy key, and employs a dynamic non–linear transformation in multi-layered encryption, which is AES-256-bit encrypted. This comprehensive approach is designed to safeguard the security of satellite data in transit from modern cryptanalysis as it traverses communication networks. The approach begins with entropy-based initializations, iteratively expands the key space, and applies adaptive transformations to render the encrypted data highly unpredictable. Experiments with satellite images of various resolutions demonstrate that this approach is resistant to cryptanalysis. The evaluation included measuring entropy, analysing the correlation between neighbouring pixels, and testing resistance to statistical and frequency-based attacks. The encrypted images achieved entropy values close to the theoretical maximum and showed almost no correlation between adjacent pixels, demonstrating the strength and uniformity of the encryption process. Performance tests on systems with multiple threads and processors revealed clear links between execution time, data size, and the level of parallelization. Moderate parallelism provided the best speed improvements, and the method remained scalable for large datasets, making it suitable for high-throughput environments. This approach ensures strong satellite-image robustness in image size, content, or spectral characteristics. The flexible structure and good performance make it a promising candidate for future telecom networks and secure satellite data distribution systems.
A Hybrid Neural Network Architecture for Semantic-Contextual Analysis of Emotions in Social Media
(CEUR-WS Team, Aachen, Germany, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Ovcharenko, Maksym; Ivanko, Artem
ENG: The paper proposes a hybrid neural network architecture for multi-class classification of the emotional state of social media texts, which combines contextual vector representations generated by the BERT model with extended structured features related to the user and message metadata. The proposed approach uses a multilayer perceptron network that combines linguistic and contextual information in a single representation. The results of the experimental study confirm the superiority of the proposed approach over traditional LSTM and CNN architectures, as well as over the separate use of BERT embeddings. The achieved classification accuracy is 90%, and the F1-measure is 0.91, which indicates the high efficiency of the model in conditions of high variability of language structures and stylistic features of social content.
Hybrid Quantum CNN-Based Information Technology for Building Semantic Segmentation in Aerial Imagery
(CEUR-WS Team, Aachen, Germany, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Babets, Dmytro; Cyran, Krzysztof; Wereszczyński, Kamil
ENG: This paper presents information technology for semantic segmentation of building objects in high-resolution aerospace images using a hybrid quantum convolutional neural network (QCNN). Quantitative and visual evaluations demonstrate the effectiveness of the proposed approach compared to classical segmentation models such as U-Net, CNN, and FCN. The hybrid QCNN model achieved high pixel classification accuracy, with a mean accuracy of 0.98 on the training set and 0.97 on the validation set, indicating robust object recognition. Loss values reached minimal levels (0.05 training, 0.07 validation), confirming efficient training without overfitting. Intersection over Union (IoU) scores were 0.90 and 0.89 for training and validation sets, respectively, demonstrating precise building contour delineation. Testing on diverse urban scenes yielded 100% detection accuracy without false positives or negatives. Training dynamics showed rapid convergence, with mean average precision (mAP) increasing to 0.35–0.45 and mAP@50:95 reaching 0.20–0.30, stabilizing after 50 epochs. The proposed technology effectively segments buildings under varying density, geometric complexity, shadows, and background noise. The hybrid QCNN approach enhances segmentation quality and generalization to new images. The results confirm the practical applicability of the developed technology for geospatial monitoring, urban planning, and mapping based on aerospace imagery.
Intelligent Sentinel Satellite Image Processing Technology for Land Cover Mapping
(Dnipro University of Technology, Dnipro, 2024) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Laktionov, I. S.; Diachenko, H. H.
ENG: Purpose. This article proposes to develop an intelligent Sentinel satellite image processing technology for land cover mapping using convolutional neural networks. The result will be an image with improved spatial resolution. Methodology. The paper presents a technology using a combination of biquadratic interpolation, histogram alignment, PCA transform, as well as a parallel residual architecture of convolutional neural networks. The technology increases the information content of Sentinel-2 optical images by combining 10 and 20-meter resolution data, resulting in primary 20-meter images with improved spatial resolution. Findings. The root mean square error (RMSE = 3.64) indicates a high accuracy in reproducing the spectral properties of the images. The correlation coefficient (CC = 0.997) confirms a high linear relationship between the estimated and observed images. The low value of Spectral Angle Mapper (SAM = 0.52) with the high Universal Image Quality Index (UIQI = 0.999) indicates high quality and structural similarity between the synthesized and reference images. These results confirm the proposed technology’s effectiveness in enhancing the spatial resolution of Sentinel satellite images. Originality. Traditional pansharpening methods of multispectral images developed for satellite images with panchromatic channels cannot be directly applied to Sentinel multispectral data, because these images do not contain a panchromatic channel. In addition, atmospheric conditions and the presence of clouds affect the quality of optical images, complicating their further thematic processing. The proposed technology, using biquadratic interpolation, histogram alignment, convolutional neural networks, and PCA transformation, removes clouds and enhances the spatial resolution of the primary 20-meter optical satellite image channels of Sentinel-2. This technology reduces color distortion and increases the detail of digital optical images, which allows for more accurate analysis of the state of the earth’s surface. Practical value. The results obtained can be used to improve the methods for processing Sentinel satellite images, which provide high spatial resolution and accurate preservation of spectral characteristics. It provides the foundation for the development of new geographic information systems for land cover monitoring.
Intelligent Technology for Land Cover Monitoring due to Amber Mining on Optical Satellite Images
(Dnipro University of Technology, Dnipro, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.
ENG: Purpose. This article proposes to develop an intelligent technology for detecting land cover changes due to amber mining based on Sentinel-2 optical satellite images. Methodology. The presented intelligent technology combines methods of geometric and radiometric correction, the Dark Object Subtraction algorithm, a hybrid architecture of convolutional neural networks (CNN + EfficientNet-Edge), and an algorithm for detecting changes over time based on the symmetric difference of changed pixels, which ensures high accuracy and efficiency in the process of analysing land cover changes. Findings. The effectiveness of the proposed technology was assessed using the F1-score, Recall, Precision, and Accuracy metrics. The values of the recall and F1-score metrics (92 %) confirm the ability of the system to detect the boundaries of land cover disturbance zones accurately. In addition, the accuracy (94 %), recall (90 %), and overall accuracy (93 %) confirm the ability of the model to effectively classify both mining-impacted areas and areas without signs of anthropogenic interference with a minimum number of errors. The low value of land cover change segmentation error (4.7 %) indicates the high quality of spatial interpretation of the results. Originality. The paper proposes a comprehensive use of convolutional neural networks with the EfficientNet-Edge architecture to detect land cover changes due to amber mining. This approach overcomes the limitations of traditional feedforward neural networks associated with problems of incorrect initialization and suboptimal distribution of weight coefficients. In particular, a comprehensive use of two feature processing levels is proposed: the first one processes simple texture features obtained using average pooling in the CNN architecture, and the second one processes spectral features formed from a 4D tensor at the last convolutional layer of EfficientNet-Edge. This helps to overcome the instability of the training process and improve the ability of the model to detect land cover changes caused by hydromechanical amber mining accurately. An annotation of areas resulting from amber mining and areas without signs of anthropogenic interference was developed. Practical value. The developed technology has practical value for determining changes in the land cover caused by hydromechanical extraction of amber. The theoretical results obtained allow for an effective assessment of the scale of changes and facilitate informed management decisions in environmental protection.
Machine Learning for Automatic Extraction of Water Bodies Using Sentinel-2 Imagery
(National University "Zaporizhzhia Polytechnic", Zaporizhzhia, 2024) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.
ENG: Context. Given the aggravation of environmental and water problems, there is a need to improve automated methods for extracting and monitoring water bodies in urban ecosystems. The problem of efficient and automated extraction of water bodies is becoming relevant given the large amount of data obtained from satellite systems. The object of study is water bodies that are automatically extracted from Sentinel-2 optical satellite images using machine learning methods. Objective. The goal of the work is to improve the efficiency of the process of extracting the boundaries of water bodies on digital optical satellite images by using machine learning methods. Method. The paper proposes an automated information technology for delineating the boundaries of water bodies on Sentinel-2 digital optical satellite images. The process includes eight stages, starting with data download and using topographic maps to obtain basic information about the study area. Then, the process involved data pre-processing, which included calibrating the images, removing atmospheric noise, and enhancing contrast. Next, the EfficientNet-B0 architecture is applied to identify water features, facilitating optimal network width scaling, depth, and image resolution. ResNet blocks compress and expand channels. It allows for optimal connectivity of large-scale and multi-channel links across layers. After that, the Regional Proposal Network defines regions of interest (ROI), and ROI alignment ensures data homogeneity. The Fully connected layer helps in segmenting the regions, and the Fully connected network creates binary masks for accurate identification of water bodies. The final step of the method is to analyze spatial and temporal changes in the images to identify differences, changes, and trends that may indicate specific phenomena or events. This approach allows automating and accurately identifying water features on satellite images using machine learning. Results. The implementation of the proposed technology is development through Python software development. An assessment of the technology’s accuracy, conducted through a comparative analysis with existing methods, such as water indices and K-means, confirms a high level of accuracy in the period from 2017 to 2023 (up to 98%). The Kappa coefficient, which considers the degree of consistency between the actual and predicted classification, confirms the stability and reliability of our approach, reaching a value of 0.96. Conclusions. The experiments confirm the effectiveness of the proposed automated information technology and allow us to recommend it for use in studies of changes in coastal areas, decision-making in the field of coastal resource management, and land use. Prospects for further research may include new methods that seasonal changes and provide robustness in the selection and mapping of water surfaces.
A Neural Network Approach to Semantic Segmentation of Vehicles in Very High Resolution Images
(National University "Zaporizhzhia Polytechnic", Zaporizhzhia, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Udovyk, Iryna M.; Kazymyrenko, O. V.; Radionov, Y. D.
ENG: Context. The semantic segmentation of vehicles in very high resolution aerial images is essential in developing intelligent transportation systems. It allows for the automation of real-time traffic management and the detection of congestion and emergencies. Objective. This work aims to develop and evaluate the effectiveness of a neural network approach to semantic segmentation in very high resolution aerial images, which provides high detail and correct reproduction of object boundaries. Method. The DeepLab architecture with ResNet-101 as a backbone is used for gradient preservation and multiscale feature analysis. We trained on DOTA data and retrained on specialized sets with classes: vehicles, green areas, buildings, and roads. A loss function based on the Dice coefficient was applied to reduce the imbalance of classes. It effectively solves the class imbalance problem and improves the accuracy of segmenting objects of different sizes. Using ResNet-101 instead of Xception in the backbone network allows us to maintain the gradient as the network depth increases. Results. Experimental studies have confirmed the effectiveness of the proposed approach, which achieves a segmentation accuracy of more than 90%, outperforming existing analogs. The use of multiscale feature analysis allows for preserving the texture features of objects, reducing false classifications. A comparative study with U-Net, SegNet, FCN8s, and other methods confirms the higher performance of the proposed approach in terms of mIoU (82.3%) and Pixel Accuracy (95.1%). Conclusions. The experiments confirm the effectiveness of the proposed method of semantic segmentation of vehicles in ultrahigh spatial resolution images. Using DeepLab v3+ResNet-101 significantly improves the quality of vehicle segmentation in an urbanized environment. Excellent metric performance makes it promising for infrastructure monitoring and traffic planning tasks. Further research will focus on adapting the model to new datasets.
Quantum Machine learning for Fusion of Multichannel Optical Satellite Images
(Dnipro University of Technology, Dnipro, 2025) Kashtan, Vita Yu.; Hnatushenko, Volodymyr V.; Wereszczyński, Kamil; Cyran, Krzysztof
ENG: Purpose. To develop a novel approach for fusion of optical satellite images based on machine learning and quantum optimization for integrating spatial-spectral information from RGB and IR channels. Methodology. The proposed approach involves sequential processing of input data, including geometric, radiometric, and atmospheric corrections. Each channel is decomposed into low-frequency and high-frequency components using a Gaussian filter. The Independent Component Analysis (ICA) method is applied to reduce the dimensionality of input data. A quantum optimizer approximation algorithm is applied to analyze the infrared channel. A deep convolutional neural network with residual dense blocks is used to extract spatial structural features from RGB channels. After integrating features through fully connected layers, the quantum block optimizes the weight coefficients for the final channel fusion. Findings. Quantitative evaluation demonstrates that the proposed approach outperforms classical fusion methods, including Brovey, Gram-Schmidt, IHS, HCS, HPFC, ATWT, PCA, and CNN, in spectral and spatial information integration accuracy. The method achieves the lowest mean squared error (MSE = 191.8), high structural similarity index (SSIM = 0.43), entropy (Entropy = 7.54), and Sobel filter range (Sobel Sharp = 19.19–21.67 across R, G, B channels). Visual analysis also confirms qualitative advantages: images exhibit clear structure without artifacts and balanced color reproduction consistent with the spectral characteristics of the original RGB data. Originality. A novel approach to utilizing information of the IR channel is proposed, which integrates a quantum-classical algorithm within a deep convolutional neural network architecture for synergistic processing of multichannel optical images using multilevel frequency decomposition and a weighted feature fusion mechanism. Practical value. The proposed approach can be implemented in Earth remote sensing systems to enhance the quality of satellite image processing, particularly for mapping, land resource monitoring, agricultural control, and environmental analysis tasks. Applying quantum algorithms opens new opportunities for improving efficiency and accuracy in processing multidimensional geoinformation data containing IR channel information.