Ai-Powered Super-Resolution Techniques for Satellite Imaging

Abstract

The escalating intricacy of worldwide air traffic management necessitates novel surveillance technologies that surpass conventional radar systems. This chapter examines the use of artificial intelligence (AI) and machine learning (ML) in the analysis of satellite images to improve air traffic surveillance. The proposed AI framework employs satellite remote sensing, computer vision techniques, and geo-referenced aircraft data to enhance real-time detection and categorization. It mitigates deficiencies in traditional systems, especially in regions devoid of radar coverage. The research delineates a tripartite methodology: collecting radar coverage from satellite photos, annotating data with geo-referenced aircraft positions, and using deep learning models for categorization. YOLO and Faster R-CNN models accurately differentiate airplanes from other objects. Experimental studies indicate the viability of AI-enhanced satellite surveillance, resulting in greater detection in high-traffic areas. The technology improves situational awareness, streamlines flight planning, alleviates airspace congestion, and bolsters security. It facilitates catastrophe response by allowing swift search-and-rescue operations. Challenges such as inclement weather and nocturnal surveillance persist, necessitating the use of infrared sensors and radar-based methodologies. The paper presents a scalable, cost-efficient strategy for future air traffic control via the integration of big data analytics, cloud computing, and satellite surveillance. Subsequent research will enhance models and broaden predictive analytics for autonomous surveillance, transforming aviation safety and operational intelligence.