Request Paper

Abstract
Maritime border security faces growing challenges due to increasing threats such as smuggling, illegal fishing, piracy, and unauthorized vessel entry. Traditional surveillance methods often struggle with limitations in real-time detection, tracking accuracy, and scalability. This study explores the integration of AI-powered object detection and tracking systems as a transformative solution for enhancing maritime border security using a novel ship detection approach called YOLOv9 with Adan optimizer (YOLOv9-Adan). Leveraging advanced deep learning algorithms and computer vision, these systems enable accurate identification and continuous monitoring of vessels across vast oceanic areas, including under adverse weather and low-visibility conditions. Our model is trained on a drone-image dataset comprising 3200 images of maritime scenes and ship types in drone views collected from various sources. The experimental results show that our approach using the YOLOv9-Adan model achieves 65.5% mAP, which exceeds the mAP of YOLOv9 by 4.3%. Additionally, this article also provides a comparative analysis of our model YOLOv9-Adan with other existing models in literature with consistently surpassing existing approaches. Keywords: Ship Detection, Deep Learning, YOLO (You Only Look Once), Adan (ADAptive Nesterov momentum algorithm), PGI (Programmable Gradient Information), GELAN (Generalized Efficient Layer Aggregate Network).