Salvato in:
Dettagli Bibliografici
Autori principali: Sarkar, Sudipto, Hasan, Mohammad Asif, Shahriar, Khondokar Ashik, Labiba, Fablia, Tasnim, Nahian, Fattah, Sheikh Anawarul Haq
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.10765
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911206488408064
author Sarkar, Sudipto
Hasan, Mohammad Asif
Shahriar, Khondokar Ashik
Labiba, Fablia
Tasnim, Nahian
Fattah, Sheikh Anawarul Haq
author_facet Sarkar, Sudipto
Hasan, Mohammad Asif
Shahriar, Khondokar Ashik
Labiba, Fablia
Tasnim, Nahian
Fattah, Sheikh Anawarul Haq
contents Identifying drones and birds correctly is essential for keeping the skies safe and improving security systems. Using the VIP CUP 2025 dataset, which provides both RGB and infrared (IR) images, this study presents EGD-YOLOv8n, a new lightweight yet powerful model for object detection. The model improves how image features are captured and understood, making detection more accurate and efficient. It uses smart design changes and attention layers to focus on important details while reducing the amount of computation needed. A special detection head helps the model adapt to objects of different shapes and sizes. We trained three versions: one using RGB images, one using IR images, and one combining both. The combined model achieved the best accuracy and reliability while running fast enough for real-time use on common GPUs.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10765
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EGD-YOLO: A Lightweight Multimodal Framework for Robust Drone-Bird Discrimination via Ghost-Enhanced YOLOv8n and EMA Attention under Adverse Condition
Sarkar, Sudipto
Hasan, Mohammad Asif
Shahriar, Khondokar Ashik
Labiba, Fablia
Tasnim, Nahian
Fattah, Sheikh Anawarul Haq
Computer Vision and Pattern Recognition
Identifying drones and birds correctly is essential for keeping the skies safe and improving security systems. Using the VIP CUP 2025 dataset, which provides both RGB and infrared (IR) images, this study presents EGD-YOLOv8n, a new lightweight yet powerful model for object detection. The model improves how image features are captured and understood, making detection more accurate and efficient. It uses smart design changes and attention layers to focus on important details while reducing the amount of computation needed. A special detection head helps the model adapt to objects of different shapes and sizes. We trained three versions: one using RGB images, one using IR images, and one combining both. The combined model achieved the best accuracy and reliability while running fast enough for real-time use on common GPUs.
title EGD-YOLO: A Lightweight Multimodal Framework for Robust Drone-Bird Discrimination via Ghost-Enhanced YOLOv8n and EMA Attention under Adverse Condition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.10765