B-BACN: Bayesian Boundary-Aware Convolutional Network for Crack Characterization
This work addresses the challenge of reliable crack characterization for structural health monitoring, representing an incremental improvement with a novel method for uncertainty quantification.
The paper tackled the problem of uncertainty quantification in crack boundary detection for structural health monitoring by proposing a Bayesian Boundary-Aware Convolutional Network (B-BACN), which concurrently quantifies epistemic and aleatoric uncertainties and improves detection precision, as demonstrated with benchmark results showing minimized misclassification rates and enhanced model calibration.
Accurately detecting crack boundaries is crucial for reliability assessment and risk management of structures and materials, such as structural health monitoring, diagnostics, prognostics, and maintenance scheduling. Uncertainty quantification of crack detection is challenging due to various stochastic factors, such as measurement noises, signal processing, and model simplifications. A machine learning-based approach is proposed to quantify both epistemic and aleatoric uncertainties concurrently. We introduce a Bayesian Boundary-Aware Convolutional Network (B-BACN) that emphasizes uncertainty-aware boundary refinement to generate precise and reliable crack boundary detections. The proposed method employs a multi-task learning approach, where we use Monte Carlo Dropout to learn the epistemic uncertainty and a Gaussian sampling function to predict each sample's aleatoric uncertainty. Moreover, we include a boundary refinement loss to B-BACN to enhance the determination of defect boundaries. The proposed method is demonstrated with benchmark experimental results and compared with several existing methods. The experimental results illustrate the effectiveness of our proposed approach in uncertainty-aware crack boundary detection, minimizing misclassification rate, and improving model calibration capabilities.