Tableware Ceramics Defect Detection Using Morphological Operation Approach

Inspections on product quality need to be carried out by manufacturing companies to ensure and maintain customer satisfaction. Quality control of physical products can be done using a digital image processing approach. Previous research conducted product defect detection using filtering and edge detection techniques based on image acquisition on flat objects. In some cases, image acquisition is taken on non-flat objects and has a background. The object's region of interest (ROI) needs to be analyzed to get the required object area. This paper aims to detect the location of the defect in the product using a morphological approach. The object of this research is a plain ceramic tableware plate consisting of a plate that has defect and has no defect. A morphological approach is used to segment the image to get the ROI and detect the location of the defect on the plate. Dilation, erosion, opening, and closing operators are utilized to obtain the expected results. The results in this study showed an accuracy of 85.76%.