Adaptive Signal Analysis for Automated Subsurface Defect Detection Using Impact Echo in Concrete Slabs
Deepthi Pavurala, Duoduo Liao, Chaithra Reddy Pasunuru·December 23, 2024
Summary
A novel automated method for detecting subsurface defects in concrete slabs using Impact Echo signal analysis is presented. The approach integrates advanced signal processing, clustering, and visual analytics. Adaptive frequency thresholding identifies anomalies tailored to each slab's distinct material properties. The methodology generates frequency maps, binary masks, and cluster maps for automatic classification. Evaluations using a labeled dataset of eight reinforced concrete specimens with known defects show high performance metrics, with F1-scores up to 0.95 and AUC-ROC of 0.83. The results demonstrate robustness, identifying defect-prone areas with minimal false positives. Adaptive frequency thresholding ensures scalability and adaptability to other frequency-based signals. This automated pipeline minimizes manual intervention, enhancing structural health monitoring. The approach holds potential for integrating multimodal sensor fusion in infrastructure maintenance.
Advanced features