Character Recognition And Printing Quality Inspection Of Chips Under Packaging Film

At present, tape reel packaging is widely used in semiconductor industry for the packaging of finished chips. Most of the incoming material inspection of tape reel packaging chips are traditional human visual inspection, which is slow and subjective. In this paper, machine vision technology is used to replace human visual inspection, the objective and accurate real-time inspection of the character recognition and the printing quality of chips under packaging film are realized. Using of polarized light source can effectively suppress the reflection of packaging film and realized high-quality image acquisition of the chip under packaging film; A characters OCR recognition technique based on MSER and deep learning is proposed, taking into account the speed and accuracy of character recognition; The high quality characters regional template is obtained by multi-image training and the morphological regional comparison method is adopted, which can effectively cope with the fluctuation of character grayscale and font thickness within a reasonable range and improve the accuracy of character printing quality inspection.