Segment anything model-based crack segmentation using low-rank adaption fine-tuning

Abstract

High-precision crack segmentation is crucial for analyzing and maintaining the apparent state of structures. The introduction of large vision models, such as the segment anything model (SAM), has brought significant advancements in object segmentation due to their remarkable generalization capabilities. However, SAM cannot be directly used for the purpose of automatic crack segmentation. This study introduces a novel approach that fine-tunes SAM specifically for crack segmentation by incorporating low-rank adaptation (LoRA). This method involves adding a dedicated crack segmentation head to SAM, enabling automatic crack segmentation. Additionally, the application of LoRA technology facilitates the readjustment of SAM’s features without incurring the substantial costs typically associated with fine-tuning entire networks. A comparative analysis with current leading crack segmentation models demonstrated a significant increase in accuracy across eight different crack datasets. This study offers guidelines for the application of large vision models for crack identification.