SpectralCA: Hybrid Neural Network for UAV-Based Hyperspectral Landmine Detection
Вантажиться...
Дата
2025
Автори
Науковий керівник
Назва журналу
Номер ISSN
Назва тому
Видавець
Igor Sikorsky Kyiv Polytechnic Institute
Анотація
Master's dissertation: 111 p., 29 figures, 20 tables, 41 references, and
2 appendices.
The relevance of this research lies in the growing demand for unmanned
aerial vehicles (UAVs) capable of operating reliably in complex environments where
conventional navigation becomes unreliable due to interference, poor visibility, or
camouflage. Hyperspectral imaging (HSI) provides unique opportunities for
UAV-based computer vision by enabling fine-grained material recognition and object
differentiation, which are critical for navigation, surveillance, agriculture, and
environmental monitoring.
The object of the study is hyperspectral image classification.
The subject of the research is development of a high-performance deep
learning architecture for a classification task in hyperspectral imaging.
The purpose of this work is to modify the Mobile 3D Vision Transformer
(MDvT) by introducing the proposed SpectralCA block. This block employs
bi-directional cross-attention to fuse spectral and spatial features, enhancing accuracy
while reducing parameters and inference time. Experimental evaluation was
conducted on the WHU-Hi-HongHu dataset, with results assessed using Overall
Accuracy, Average Accuracy, and the Kappa coefficient.
The findings confirm that the proposed architecture improves UAV perception
efficiency, enabling real-time operation for navigation, object recognition, and
environmental monitoring tasks.
Versions of this work were published in arXiv and presented at the 2025 IEEE
8th International Conference on Methods and Systems of Navigation and Motion
Control. Also, this research won first place at the 2025 International Student Research
Competition in Artificial Intelligence.
Опис
Ключові слова
SpectralCA, deep learning, computer vision, hyperspectral imaging, unmanned aerial vehicle, object detection, semi-supervised learning, SpectralCA, глибоке навчання, комп'ютерний зір, гіперспектральне бачення, безпілотний літальний апарат, виявлення об'єктів, напівкероване навчання
Бібліографічний опис
Brovko, D. V. SpectralCA: Hybrid Neural Network for UAV-Based Hyperspectral Landmine Detection : Master’s Thesis : 122 Computer Sciences / Brovko, Danil Vitaliiovych. – Kyiv, 2025. – 114 p.