Defect measurement in CFRP drilling based on digital image processing

Drilling-induced burr and delamination are the main defects in carbon fiber–reinforced polymers (CFRPs). Although substantial CFRP defect measurement methods have been suggested, it is still challenging to recognize and measure the defects due to the drilling-induced chips, non-drilling-induced material reflections, and textured surface of CFRPs. To this end, the main purpose of the present paper is to develop a method able to recognize and measure the drilling-induced burr and delamination in CFRPs based on two techniques of bimodal threshold segmentation method and K-means clustering. To implement these techniques, images are captured by the image capturing setup with specific illumination. And the proposed defect measurement method is validated through full factorial drilling experiments in CFRP workpieces and compared with manual measurement supported by conventional image processing method. The results of the comparative measurement on the proposed method and manual method show that there is a reasonable agreement between them. Moreover, the proposed method can separate burr and delamination from regions of chips and textured surface of CFRPs. Considering the efficient, automated, and repeatable nature of the proposed method, it is anticipated to provide a meaningful reference for recognizing and measuring drilling-induced defects in the presence of strong interfering targets.