Defect Detection of Skewed Images for Multilayer Ceramic Capacitors

In this paper, we utilized machine vision and image processing to develop an image detection flow for the dimension and appearance of multilayer ceramic capacitors (MLCC), and also used proposed automatic optical inspection (AOI) system in the MLCC production line operation. We compared the advantages and disadvantages of Hough Transform and Histogram analysis. The proposed tiny passive components detection flow can execute the defect detection of each capacitor in real time as soon as the image information of the component is obtained. If defect can be found on the detection flow, it can be immediately judged and classified into defective components, and detection time of each component can be significantly reduced.