Vision-Based Structural Health Monitoring System Using Edge Computing Device

Abstract

As the development of various IoT devices are continuously expanding, the vision-based systems are gradually increasing in the field of Structural Health Monitoring (SHM) system. This paper focuses on the development of a computational algorithm to measure the frequency of structural beams from vision image data. The image processing algorithms are contained in a small IoT device, Raspberry Pi 4 and compatible HQ camera. The device analyzes the changes in the frequency in accordance with the variations in in the bolt looseness. For the accurate analysis, image processing techniques using OpenCV, Fourier transform, and mathematical model have been used. The result of the research shows the great potential for real-world applications, which is expected to save a lot of cost and effort.