As traditional ship crack detection methods based on artificial visual inspection and ultrasonic methods in ship repair and inspection processes have the characteristics of low efficiency, high cost and high danger, a ship crack detection method based on deep learning is proposed.

First, a lightweight convolutional structure (GSConv) is used to replace the standard convolution and introduce an attention mechanism in the backbone of YOLOv5s to achieve the reduction of network parameters and computational complexity while enhancing the ability to extract crack features. Second, C3_Faster constructed by a fast convolutional structure is used instead of the original C3 module in the neck of the network to improve the processing speed of the model and enhance its rapidity. Finally, an improved bidirectional weighted feature fusion network (BiFFN) is used to enrich the semantic and positional information of cracks in the feature map and improve the model's crack identification accuracy and location precision.

By training on both original and augmented ship crack datasets, the proposed method achieves a detection accuracy of over 94.11% and a recall rate of over 93.50%, while reducing the computational complexity by 17.93% and parameter count by 15.81%.

This study demonstrates that the proposed ship crack detection method based on GSConv and BiFFN achieves lightweight model architecture and high detection accuracy and recall rates, providing useful references for the development of UAV/ship autonomous inspection systems.