Noncontact Sensing Techniques for AI-Aided Structural Health Monitoring: A Systematic Review

Abstract

Engineering structures and infrastructure continue to be used despite approaching or having reached their design lifetime. While contact-based measurement techniques are challenging to implement at a large scale and provide information at discrete locations only, noncontact methods are more user-friendly and offer accurate, robust, and continuous spatial information to quantify the structural conditions of the targeted systems. Advancements in optical sensors and image-processing algorithms increased the applicability of image-based noncontact techniques, such as photogrammetry, infrared thermography, and laser imaging for structural health monitoring (SHM). In addition, with the incorporation of artificial intelligence (AI) algorithms, the assessment process is expedited and made more efficient. This article summarizes the efforts made in the last five years to leverage AI-aided noncontact sensing techniques for applications in SHM with an emphasis on image-based methods. Future directions to advance AI-aided image-based sensing techniques for SHM of engineering structures are also discussed.