A Study of Suitable LED Shading for Fiducial Marker Detection on PCB Images
Conference proceedings article
Authors/Editors
Strategic Research Themes
Publication Details
Author list: CHANYARAT KASWSAVATH, APICHAI BHATRANAND, YUTTAPONG JIRARAKSOPAKUN
Publication year: 2021
Abstract
In PCB assembly lines, lighting is crucial for in-line object detection in machine vision system. Insufficient lighting can decrease image quality or object sharpness, causing a disruption of PCB assembly. This study investigated an effective LED condition for detecting copper circular fiducial marks on PCB images by a Besi DataCon 2200 machine with 6 shades of LED at 9 various light intensities from 0-100. The experimental results demonstrate that using a yellow LED at 10% intensity level can raise the detection rate of the fiducial marks to 98% and increase CNR ratio by 9.871%, resulting in a reduction in downtime for the Besi DataCon 2200 machine during PCB assembly procedures.
Keywords
No matching items found.