Efficient Detection of Runways, Taxiways, and Aprons in Commercial Airports Using YOLOv8 and High-Resolution Satellite Imagery

This study presents an advanced satellite-based detection system designed for the comprehensive identification of multiple facilities within both commercial and civilian airports. The system utilizes cutting-edge remote sensing technologies and machine learning algorithms to accurately detect and categorize various essential elements crucial for airport operations. These elements include runways, taxiways, and aprons. Using high-resolution satellite imagery, the system aims to improve airport infrastructure management, optimize safety protocols, and improve overall operational efficiency on a global scale. Leveraging deep learning techniques, specifically object detection models, we explore the suitability of different architectures such as Faster R-CNN, YoLO V8, detectors for accurately identifying these facilities. The project involves extensive data collection, preprocessing, model training, and evaluation to ensure robust performance. Considering the potential deployment on edge devices, the chosen models are optimized for efficiency without compromising accuracy.