OVERVIEW OF COMPUTER VISION ALGORITHMS FOR DETECTING HAZARDOUS OBJECTS BY DRONES
DOI:
https://doi.org/10.26906/SUNZ.2023.3.120Keywords:
computer vision, drone, neural networks, demining, reconstruction, combat operationsAbstract
A theoretical overview of classification algorithms has been conducted to construct relevant research models for the detection of hazardous objects. The formal problem statement for the classification of hazardous objects has been formulated to build classification models. This task is addressed by constructing classification models using machine learning tools or neural networks. The complexity of the research lies in selecting a neural network architecture tailored for image classification, which will be used as input. The graphical representation of the neural network architecture is implemented using the networks and matplotlib libraries. This allows for a comprehensive understanding of the overall task in creating classification models. The models will be used in future research as one of the functions of a drone for detecting dangerous objects in combat conditions.Downloads
References
Soori M. Artificial Intelligence, Machine Learning and Deep Learning in Advanced Robotics, A Review [Electronic resource] / Mohsen Soori, Behrooz Arezoo, Roza Dastres // Cognitive Robotics. – 2023. – Mode of access: https://doi.org/10.1016/j.cogr.2023.04.001 (date of access: 27.08.2023). – Title from screen.
Panibratov R. Decision support system for estimating and forecasting state of insurance company [Electronic resource] / Roman Panibratov // System research and information technologies. – 2022. – No. 1. – P. 61–72. – Mode of access: https://doi.org/10.20535/srit.2308-8893.2022.1.05 (date of access: 20.08.2023). – Title from screen.
Multi-task learning for dangerous object detection in autonomous driving [Electronic resource] / Yaran Chen [et al.] // Information Sciences. – 2018. – Vol. 432. – P. 559–571. – Mode of access: https://doi.org/10.1016/j.ins.2017.08.035 (date of access: 27.08.2023). – Title from screen.
Improving handgun detection through a combination of visual features and body pose-based data [Electronic resource] / Jesus Ruiz-Santaquiteria [et al.] // Pattern Recognition. – 2022. – P. 109252. – Mode of access: https://doi.org/10.1016/j.patcog.2022.109252 (date of access: 27.08.2023). – Title from screen.
Report on UG2+ challenge Track 1: Assessing algorithms to improve video object detection and classification from unconstrained mobility platforms [Electronic resource] / Sreya Banerjee [et al.] // Computer Vision and Image Understanding. – 2021. – Vol. 213. – P. 103297. – Mode of access: https://doi.org/10.1016/j.cviu.2021.103297 (date of access: 27.08.2023).
Towards a framework of key technologies for drones [Electronic resource] / Réda Nouacer [et al.] // Microprocessors and Microsystems. – 2020. – Vol. 77. – P. 103142. – Mode of access: https://doi.org/10.1016/j.micpro.2020.103142 (date of access: 27.08.2023). – Title from screen.
Dynamic computation offloading for ground and flying robots: Taxonomy, state of art, and future directions [Electronic resource] / Rihab Chaâri [et al.] // Computer Science Review. – 2022. – Vol. 45. – P. 100488. – Mode of access: https://doi.org/10.1016/j.cosrev.2022.100488 (date of access: 27.08.2023). – Title from screen.
Lee S. W. Decision-making in brains and robots – the case for an interdisciplinary approach [Electronic resource] / Sang Wan Lee, Ben Seymour // Current Opinion in Behavioral Sciences. – 2019. – Vol. 26. – P. 137–145. – Mode of access: https://doi.org/10.1016/j.cobeha.2018.12.012 (date of access: 27.08.2023). – Title from screen.
Аналіз систем розпізнавання образів структури композитів : монографія / Добротвор І.Г., Стухляк П.Д., Микитишин А.Г., Митник М.М. – Тернопіль : Тернопільський національний технічний університет імені І. Пулюя, 2018. – 224 с.
Кононова К. Ю. Машинне навчання: методи та моделі: підручник для бакалаврів, магістрів та докторів філософії спеціальності 051 «Економіка» / К. Ю. Кононова. – Харків: ХНУ імені В. Н. Каразіна, 2020. – 301 с.
Мацуга, О.М. Навчальний посібник до вивчення курсу «Інформаційні технології розпізнавання образів» [Текст] / О.М. Мацуга, Ю.М. Архангельська, Н.М. Єрещенко. – Д.: РВВ ДНУ, 2016. – 60 с.
European Environment Agency's home page [Electronic resource] // European Environment Agency's home page. – Mode of access: https://www.eea.europa.eu/en (date of access: 27.08.2023). – Title from screen.
Субботін С. О. Нейронні мережі: теорія та практика: навч. посіб. / С. О. Субботін. – Житомир: Вид. О. О. Євенок, 2020. – 184 с
Терейковський І. Штучні нейронні мережі: базові положення [Електронний ресурс] / І. Терейковський, Д. Бушуєв, Л. Терейковська. – Режим доступу: https://ela.kpi.ua/bitstream/123456789/50135/1/ANN.pdf (дата звернення: 27.08.2023).
DiGraph - Directed graphs with self loops – NetworkX 2.0.dev20161129121305 documentation [Electronic resource] // NetworkX documentation – NetworkX 2.0.dev20161129121305 documentation. – Mode of access: https://pelegmnetworkx.readthedocs.io/en/latest/reference/classes.digraph.html (date of access: 27.08.2023). – Title from screen.
