The rapid advancement of Synthetic Aperture Radar (SAR) technology due to its ability to obtain high-resolution two-dimensional images in all-weather, all-day conditions and its strong penetration has led to its widespread applications in environmental monitoring, disaster management, maritime surveillance and target detection. Ship target detection in SAR imagery plays a critical role in defense and surveillance, enabling the tracking of enemy vessels, which directly impacts military operations. In this context, the ship detection model was developed using the YOLOv10 deep learning framework, known for its robust feature extraction, high accuracy, fast processing and real-time analysis capabilities essential for handling challenging operational environments. The comparative analysis was conducted between YOLOv10 and its variants, RT-DETR and Faster R-CNN using the Multi Resolution Satellite based Ship Detection (MRSSD) dataset. YOLOv10 outperformed the other models on aspects such as detection accuracy and speed, achieving a high mAP of 0.962 and mAP@ 0.95 of 0.642. The model demonstrated high precision (0.935) and recall (0.923), effectively identifying and localizing most ships near shores with minimal false positives or false negatives. Inference speeds ranging from 1 to 21 ms on various variants of YOLOv10, makes the model suitable for real-time applications like maritime surveillance, rescue operations and ship monitoring.