Ubiquitous Robot for Structural Studies: Comparative Analysis on Concrete Crack Images using Feature Detection and Matching

The characteristic of any device to be mobilized and autonomous can bring any technology to maximize its purpose. Structural inspection plays an important function in the industry when comes to the evaluation of a building's reliability and safety. It will be more beneficial for any structural engineer or building specialist to be aided with an independent robotic device capable of performing similar roles. In this publication, it will aim to develop a system that can investigate and compare captured images of concrete cracks using feature detection and matching algorithms attached on a Level 4 (High automation) vehicle. It will identify significant key points between pre-trained concrete crack dataset from real-time captured images. Through the experimental prototype, the device was designed to be capable of performing concrete inspections. Results generated were compared based on factors applied to images such as scaling, rotation, and skewing. Overall, the implemented algorithms showed differences in terms of optimized performance like computation time and detected key points.