ПРИМЕНЕНИЕ МЕТОДОВ МАШИННОГО ОБУЧЕНИЯ ДЛЯ БОРТОВОГО РАСПОЗНАВАНИЯ ОБЪЕКТОВ И ОЦЕНКИ РАССТОЯНИЯ НА БЕСПИЛОТНЫХ ЛЕТАТЕЛЬНЫХ АППАРАТАХ
DOI:
https://doi.org/10.52167/1609-1817-2025-139-4-435-447Ключевые слова:
БПЛА, машинное обучение, компьютерное зрение, обнаружение объектов, оценка расстояния, инерциальные датчики, обработка изображенийАннотация
В данной работе рассматривается разработка архитектуры виртуальной бортовой системы БПЛА, оснащенный вычислительным модулем на базе NVIDIA Jetson Nano, RGB-камеры, IMU-сенсор и GPS-приёмником. Оно предназначено для решения задач по распознаванию целей БПЛА и определению расстояния до объектов. Для решения задач по обнаружению объектов в режиме реального времени применена облегченная модификация модели YOLOv2-Lite с использованием MobileNet-SSD в качестве экстрактора признаков. Определение расстояния до объекта реализуется применением на основе соотношения подобия треугольников по методу вычисления угловому размеру цели.
Библиографические ссылки
[1] Chen, Ching-Ju & Huang, Ya-Yu & Li, Yuan-Shuo & Chen, Ying-Cheng & Chang, Chuan-Yu & Huang, Yueh-Min. (2021). Identification of Fruit Tree Pests with Deep Learning on Embedded Drone to Achieve Accurate Pesticide Spraying. IEEE Access. PP. 1-1. 10.1109/ACCESS.2021.3056082.
[2] Fan, Kaizhe & Li, Qian & Li, Quanjun & Zhong, Guangqi & Chu, Yue & Le, Zhen & Xu, Yeling & Li, Jianfeng. (2024). YOLO-Remote: An Object Detection Algorithm for Remote Sensing Targets. IEEE Access. PP. 1-1. 10.1109/ACCESS.2024.3479320.
[3] Muzammul, Muhammad & Algarni, Abdul & Ghadi, Yazeed & Assam, Muhammad. (2024). Enhancing UAV Aerial Image Analysis: Integrating Advanced SAHI Techniques with Real-Time Detection Models on the VisDrone Dataset. IEEE Access. PP. 1-1. 10.1109/ACCESS.2024.3363413.
[4] Tang, Songtao & Zhang, Leiming & Liu, Xudong & Lv, Rongfu & Qin, Ruyi. (2025). RFHS-RTDETR: Multi-Domain Collaborative Network with Hierarchical Feature Integration for UAV-Based Object Detection. IEEE Access. PP. 1-1. 10.1109/ACCESS.2025.3586649.
[5] Ren, Libo & Yin, Wenxin & Wenhui, Diao & Fu, Kk & Sun, Xian. (2025). SuperMOT: Decoupling Motion and Fusing Temporal Pyramid Features for UAV Multiobject Tracking. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. PP. 1-15. 10.1109/JSTARS.2025.3563060.
[6] Qiu, Fen & Shen, Xiaojun & Zhou, Cheng & Yao, Lili. (2024). Rice Ears Detection Method Based on Multi-Scale Image Recognition and Attention Mechanism. IEEE Access. PP. 1-1. 10.1109/ACCESS.2024.3400254.
[7] Huang, Junyu & Luo, Renbo & Tan, Yuna & Wu, Zhuowen. (2024). CRE-YOLO: Efficient Jaboticaba Tree Detection on UAV Platforms. IEEE Access. PP. 1-1. 10.1109/ACCESS.2024.3520115.
[8] Lin, Huei-Yung & Tu, Kai-Chun & Li, Chih-Yi. (2020). VAID: An Aerial Image Dataset for Vehicle Detection and Classification. IEEE Access. 8. 212209-212219. 10.1109/ACCESS.2020.3040290.
[9] Yu, Chenhui & Liu, Yakui & Zhang, Wanru & Zhang, Xue & Zhang, Yuhan & Jiang, Xing. (2023). Foreign Objects Identification of Transmission Line Based on Improved YOLOv7. IEEE Access. PP. 1-1. 10.1109/ACCESS.2023.3277954.
[10] Liu, Zhiping & Yu, Yannan & Liang, Zhuohui & Han, Guodong & Lu, Yao. (2023). Defect Detection and Recognition of Crane Girder Structure Based on 3D Reconstruction of UAV. IEEE Access. PP. 1-1. 10.1109/ACCESS.2023.3311017.
[11] Hu, Zhuangli & He, Tong & Zeng, Yihui & Luo, Xiangyuan & Wang, Jiawen & Huang, Sheng & Liang, Jianming & Sun, Qinzhang & Xu, Hengbin & Lin, Bin. (2018). Fast image recognition of transmission tower based on big data. Protection and Control of Modern Power Systems. 3. 10.1186/s41601-018-0088-y.
[12] Mao, Yiming & Chen, Ming & Wei, Xianglin & Chen, Bing. (2020). Obstacle Recognition and Avoidance for UAVs Under Resource-Constrained Environments. IEEE Access. PP. 1-1. 10.1109/ACCESS.2020.3020632.
[13] Wei, Menghan & Xu, Kaijun. (2023). Multi-Task Matching Mechanism Design for UAV-assisted MEC Network with Blockchain. IEEE Access. PP. 1-1. 10.1109/ACCESS.2023.3332822.
[14] Jung, Sungkyu & Jang, Jaeyeon & Kim, Chang Ouk. A Lightweight Two-Step Detection Method for Real-Time Small UAV Detection. IEEE Access. PP. 1-1. 10.1109/ACCESS.2025.3590473.
[15] Shi, Cuiping & Miao, Fengjuan & Jin, Zhan & Xia, Ying. (2019). Target Recognition of Synthetic Aperture Radar Images Based on Matching and Similarity Evaluation Between Binary Regions. IEEE Access. PP. 1-1. 10.1109/ACCESS.2019.2948839.
[16] Azmat, Usman & Alotaibi, Saud & Almudawi, Naif & Alabduallah, Ibrahimm & Alonazi, Mohammed & Jalal, Ahmad & Park, Jeongmin. (2024). An Elliptical Modeling Supported System for Human Action Deep Recognition Over Aerial Surveillance.
[17] Zhou, Dongbo & Wang, Huan & Pang, Shiyan & Chen, Di & Yao, Huang & Yu, Jie. (2025). UAV-FAENet: Frequency-Aware and Attention-Enhanced Network for Remote Sensing Semantic Segmentation of UAV Imagery. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. PP. 1-17. 10.1109/JSTARS.2025.3593557.
[18] Tsai, Chi-Yi & Tsai, Shu-Hsiang. (2018). Simultaneous 3D Object Recognition and Pose Estimation Based on RGB-D Images. IEEE Access. PP. 1-1. 10.1109/ACCESS.2018.2808225.
[19] Bibi, Misbah & Faseeh, Muhammad & Khan, Anam & Rizwan, Atif & Waqas Khan, Qazi & Ahmad, Rashid & Kim, Do-Hyeun. (2025). A Unified Approach for Object Detection and Depth Map based Distance Estimation in Security and Surveillance Systems. IEEE Access. PP. 1-1. 10.1109/ACCESS.2025.3539310.
[20] Faseeh, Muhammad & Bibi, Misbah & Khan, Murad & Rizwan, Atif & Ahmed, Rashid & Kim, Do. (2024). Geo-Temporal Selective Approach for Dynamic Depth Estimation in Outdoor Object Detection and Distance Measurement. IEEE Access. PP. 1-1. 10.1109/ACCESS.2024.3485089.
[21] E. Chen, O. Haik and Y. Yitzhaky, "Online Spatio-Temporal Action Detection in Long-Distance Imaging Affected by the Atmosphere," in IEEE Access, vol. 9, pp. 24531- 24545, 2021, doi: 10.1109/ACCESS.2021.3057172.
Загрузки
Опубликован
Как цитировать
Выпуск
Раздел
Категории
Лицензия
Copyright (c) 2025 Турсунбаева Гулжамал, Дина Сатыбалдина , Арман Узбекбаев, Асел Нурушева, Наурас Әль Бухари
Это произведение доступно по лицензии Creative Commons «Attribution-NonCommercial-NoDerivatives» («Атрибуция — Некоммерческое использование — Без производных произведений») 4.0 Всемирная.
