Non-destructive photoacoustic imaging of metal surface defects

Abstract

The detection of metal surface defects is important in achieving the goals of product quality enhancement and manufacturing cost reduction. Identifying the defects with visual inspection is difficult, inaccurate, and time-consuming. Thus, several inspection methods using line cameras, magnetic field, and ultrasound have been proposed. However, identifying small defects on metal surfaces remains a challenge. To deal with this problem, we propose the use of photoacoustic imaging (PAI) as a new non-destructive imaging tool to detect metal surface defects. We successfully visualized two types of cracks (i.e., unclassified and seam cracks) in metal plate samples using PAI. In addition, we successfully extracted cracked edges from height-encoded photoacoustic maximum amplitude projection images using the Laplacian of Gaussian filtering method, and then, quantified the detected edges for a statistical analysis. We concluded that PAI can be useful in detecting metal surface defects reducing the defect rate and manufacturing cost during metal production.