This study investigates the application of YOLOv5 for object detection and classification of maritime vessels, utilizing a dataset with images annotated into five subclasses of ships. The model achieved a mean average precision (mAP@0.5) of 0.932, indicating effective detection capabilities, though some misclassifications were noted, particularly between cargo ships and tankers. The research highlights the importance of reliable object detection methods in both military and commercial maritime domains.