Integrating Shipping Domain Knowledge into Computer Vision Models for Maritime Transportation

Maritime transportation plays a significant role in international trade and the global supply chain. To enhance maritime safety reduce pollution to marine environment, various regulations conventions are proposed by organizations. ensure that shipping activities comply with relevant regulations, more attention has been paid surveillance. Specifically, cameras have widely equipped on shore drones capture videos of vessels. Then, computer vision (CV) methods adopted recognize specific type ships so as identify illegal activities. However, complex environments may hinder CV models from making accurate ship recognition. Therefore, this study proposes novel approach integrating domain knowledge, such features sailing speed, for recognition transportation, which can better support We also give two examples demonstrate great potential method future research