Corrosion characterization using fringe projection and surface analysis (Conference Presentation)

Corrosion, a global industrial challenge, costs an estimated $2.5 trillion annually to address, with its failure in critical components like automotive engines posing significant safety risks. Automating corrosion detection could mitigate human error and labor shortages in this field. This study utilizes fringe projection profilometry, a high-speed 3D profiling technique, to measure corrosion-induced morphological changes in automotive engine components, particularly connecting rods. By applying k-means clustering, we classify corrosion levels into low, medium, and high bands based on 3D imaging data. The proposed model achieved an accuracy of 88.57%, verified against expert visual classification, offering a promising step toward automated corrosion detection and prevention.