The role of satellite image technology in maritime surveillance and national sovereignty using the Synthetic Aperture Radar (SAR) data approach

Abstract Synthetic Aperture Radar (SAR) technology has emerged as a crucial tool in maritime surveillance and national sovereignty. This study integrates deep learning techniques with SAR imaging, utilizing YOLOv8 to enhance ship detection accuracy. The proposed methodology involves a structured approach to object detection using SAR data. It begins with the acquisition of SAR images and their annotation, followed by pre-processing steps that include extracting relevant image information, converting annotations from COCO to YOLO format, and creating ground truth based on bounding box areas. The pre-processed data is then utilized for training a YOLOv8 model. After model training, the performance is evaluated using metrics such as precision, recall, F1-score, and mean Average Precision (mAP) to ensure the model’s accuracy and reliability. This comprehensive framework aims to optimize object detection in SAR imagery by leveraging advanced techniques in data preparation and model evaluation.