AI-Driven Marine Vessel Detection Through Satellite Imagery: A Deep Learning Approach

This study explores an advanced methodology for detecting marine vessels via satellite imagery, utilizing a deep learning framework tailored for maritime surveillance. A novel convolutional neural network (CNN) architecture, optimized for varying vessel sizes and adverse conditions, achieved a detection accuracy of 94%, significantly surpassing traditional computer vision techniques. The approach integrates innovative preprocessing techniques and a robust post-processing framework to ensure reliability under challenging scenarios like partial cloud cover and rough sea states. This work demonstrates the potential of AI-driven solutions for enhancing real-world maritime operations, including national security and traffic management.