Semi-supervised crack detection using segment anything model and deep transfer learning

Abstract

Computer vision models have shown great potential in pavement distress detection. There is still challenge of low robustness under different scenarios. The model robustness is enhanced with more annotated data. However, this approach is labor-intensive and not a sustainable long-term solution. This paper proposes a semi-supervised instance segmentation method for road distress detection based on deep transfer learning. The interactive segmentation method utilizing SAM are used to enhance the production efficiency of segmentation datasets. The DCNv3 and lightweight segmentation heads are strategically designed to offset potential speed losses. The deep transfer learning method fine-tunes the pre-trained models, enhancing their competency for new tasks. The proposed model achieves comparable performance to supervised learning with fewer annotated data, accurately determining crack dimensions across varied scenarios. This paper provides an efficient and practical approach for pavement distress identification using the hybrid computer vision methodology.