Abstract
This study proposed an automatic optical inspection (AOI) technique to improve the inspection of chemical stains on solar wafers. Poly-silicon solar cell wafers were inspected for chemical stains, and the inspection was rapid and stable. The system used a laser-reflection-point-based AOI method for solar wafer chemical stain inspection. Based on the fuzzy theory, the image binarization algorithm could efficiently filter irrelevant image information, and the back-propagation method was also utilized to determine if the image was stained. The inspection algorithm integrated fuzzy theory and the back-propagation method in order to shorten the comparison time and quickly find the target. The experiment proved that the validity of the proposed method could achieve a recognition rate of 98% from among 1000 images.
Acknowledgements
This work was sponsored by the National Science Council under grant number NSC 101-2221-E-035 -039 -MY2 and the help of Mosel Vitelic Inc.