Research on lightweight detection of maritime ship targets based on iGF-YOLOv11

Abstract

To address the difficulty of effectively detecting maritime ship targets on mobile devices with limited computing resources under interference from illumination variations and wave disturbances, this paper proposes a lightweight ship detection model, termed iGF-YOLOv11, based on YOLOv11. Firstly, the GhostConv module is used to replace the standard convolution in the backbone network to reduce the model complexity and computational load; secondly, the iEMA attention module that integrates the advantages of iRMB and EMA mechanisms is introduced in the neck network to enhance the model's ability to extract key features of ships; moreover, a P2 detection head is added to improve the perception performance of small ship targets. Experimental results show that the improved model maintains high efficiency while significantly improving the detection accuracy. The proposed method achieves a balance between detection accuracy and lightweighting, providing a feasible solution for the deployment and operation of detection models on mobile devices.