Acoustic Sensor-Based Approach for Detecting Damage in Masonry Structures

This article proposes an acoustic sensor based model for classification, the task is to detect damage in masonry structures. In impact acoustic, we make use of a small metal object to tap on a structure to generate acoustic signals. In practical inspection, the presence of surface irregularities lead to abnormal multiple bounces. One audio recording device is used to save the acoustic signals; these signals are conveyed by different bond qualities. We faced a problem in Spectral characteristics due to bounces in the signal. Another critical issue that must be addressed is the extra noise that is added by real-life environments which leads to false judgment and feature mismatch as well. The classic feature-based classification methods are degraded due to noise. Feature extraction techniques are applied to the emitted signal to extract information. The training model-based inspection strategy is used to combine the various features to a Convolution Neural Network (CNN). Based on this approach, the bond integrity is recognized with increased disturbance to surface roughness as well as noise from the environment.