Research on ceramic 3D printing surface defect detection system based on machine vision

A vision-based online surface defect inspection system was developed to identify defects in the ceramic extrusion 3D printing process, consisting mainly of a ceramic 3D printer, a vision inspection platform, and a vision inspection module. By inspecting the surface during the printing process, the inspection system effectively avoids wasted material and time spent doing subsequent processes due to defects arising during the printing process. Image processing is carried out using Halcon software and the area to be detected is extracted using the Otsu method (OSTU) on the acquired images. Defects are detected using methods such as the Canny algorithm and the Fast Fourier Transform, and then the three defects are further differentiated according to feature differences. The experimental results show that the system can detect surface defects with an accuracy of 96.7% or more, which can meet the practical requirements. Using the confusion matrix with recall and accuracy as evaluation metrics for this system, the results show that the inspection system can accurately detect defects that exist in the printing process.