Summary: | 碩士 === 國立清華大學 === 工業工程研究所 === 84 === In the semiconductor industry,manual inspection is less
reliable due to the parts are becoming smaller and smaller.
Thus, it is necessary use machine vision to replace the manual
inspection.Automatic inspection has gained gradual emphasis in
the semiconductor industry. An automated inspection system of
wafer surface defects has developed in this study.This system
can inspect three main defects,which are the darker defects,the
lighter defects in the regular region and the defects on the
pads.The experimental result shows that the recognition rate
for the darker defects in the regular regon is 96.25%.The
recognition rate for lighter defects in the regular region is
98.75%.The recognition rate for the defects on the pads is
94.44%.However,the inspection time is a bit too long.
Suggestions about approachs to improve the speed of inspection
are proposed.
|