Satellite Still Image Classification Using CNN

• Published in: 2024 International Telecommunications Conference (ITC-Egypt) pp. 60 - 65
• Main Authors: El-den, B. M., Elbialy, Samar
• Format: Conference Proceeding
• Language: English
• Published: IEEE 22.07.2024
• Subjects: CNN; Convolutional neural networks; Deep learning; Image Classification; Natural languages; Object detection; Satellite still image; Satellites; Scene classification; Telecommunications
• DOI: 10.1109/ITC-Egypt61547.2024.10620576

Satellite still image plays a crucial role in various domains, such as law enforcement, disaster response, and environmental monitoring. The ability to manually identify objects and facilities within these images is often crucial for these applications. However, given the extensive geographic areas that require coverage and the limited number of analysts available, automation has become a necessity. Traditional object detection and classification algorithms have proven to be insufficiently accurate and reliable for solving this problem. Fortunately, deep learning, a branch of machine learning, offers promising solutions for automating such tasks. One particular approach that has achieved success in image understanding is the use of convolutional neural networks (CNNs). Deep learning has experienced substantial advancements in diverse domains, including computer vision and natural language processing, Despite the progress made in this area, there remains a dearth of comprehensive evaluations concerning datasets and techniques tailored specifically for scene classification using satellite still image. This research paper seeks to bridge this gap by conducting a comprehensive analysis of deep learning, its development over time, and its notable applications within the domain of satellite still image.