| Driven by the continuous development of the social economy,urban construction in China has witnessed a booming growth,resulting in a significant increase in the number of buildings.However,factors such as the construction quality,climatic conditions,and material standards contribute to the occurrence of cavitation phenomena on building exteriors.Cavitation defects on external walls not only pose a series of safety hazards but also harbor various internal risks.Therefore,the detection and determination of cavity defects in the external walls of buildings hold substantial engineering and social significance,as it aids in addressing concealed internal risks promptly.This paper presents a novel eddy current-based system for the rapid detection of cavity defects on building facades,which pose a significant threat to building safety.The system employs a BP neural network to analyze the two-dimensional impedance characteristics of eddy current signals and detect cavity defects on building facades.Initially,an embedded eddy current detection system is developed,utilizing selfcomparison eddy current coils and a composite magnetic filler to confer electromagnetic properties on the building facade insulation system for signal acquisition.The acquired signal is then transformed into an eddy current impedance map,and the extracted characteristic parameters of the impedance map serve as input for defect analysis.Subsequently,image processing techniques are employed to extract geometric features from the impedance map signal,and Fast Principal Component Analysis(FPCA)is utilized to reduce the dimensionality of the feature vectors.Finally,the reduced-dimensional feature vectors are employed as input for training a BP neural network model,facilitating the detection of cavity defects in building facades.To validate the efficacy of the proposed rapid defect detection system,a series of experiments were conducted in this study.The experimental results demonstrate that the eddy current non-destructive testing system,developed in this research,exhibits notable advantages in terms of speed,efficiency,and reliability.It effectively detects cavity defects on building facades,addressing the limitations of conventional testing methods,which fail to promptly and reliably identify such defects. |