FLUORESCENCE IMAGING FOR MACHINE VISION

Suitable illumination is a crucial aspect in the successful solution of machine vision problems. In this research we used objective image evaluation techniques and found that fluorescence imaging is superior to conventional illumination for acquiring images of integrated circuit lead bonds. This is an interesting and surprising finding, since there was no a priori reason to expect that any part of the bond would contain fluorescent components. Consequently, fluorescence imaging should be considered as an option in designing machine vision systems, especially if conventional illumination systems do not produce images of adequate quality. In this research we discovered a novel and effective method for threshold selection.