Adaptive cross-scenario few-shot learning framework for structural damage detection in civil infrastructure

Structural damage detection techniques are gaining widespread attention in construction engineering and management. However, the scarcity of structural damage samples and the cross-task transferability of existing knowledge currently limit this technique in practical applications. Therefore, this paper proposes a novel framework for structural damage detection with large scope of cross-task learning capability that incorporates Bayesian estimation and variational inference into the deep learning backbones and Bayesian weight function into the outer loop process of metalearning. Experimental results demonstrate the superiority of this method for both structural damage image classification and structural damage semantic segmentation. Compared with existing frameworks, the proposed method can alleviate the negative influence of domain bias and reduce computation time and costs due to sample labeling. This paper also discusses how the proposed framework can be used to train a model of the structural damage detection framework in extreme cases. The framework and findings presented in this paper have important theoretical and practical contributions to the literature on vision-based structural damage detection.