Defects extraction for QFN based on mathematical morphology and modified region growing

To extract defects from quad flat non-lead (QFN) package surface, a novel method based on mathematical morphology and modified region growing is proposed. Firstly, segment QFN images with 4-thresholds using the multilevel thesholding method. Secondly, according to the image level, obtain the shallow defects images and the deep defect images. Thirdly, eliminate the pixels around edge based on Canny edge detector. Then, use the mathematical morphology technique to remove noise pixels and locate the connected region of defects. Finally, apply modified region growing method to extracting the defects from QFN surface. The experiments show that the proposed method can extract defects efficiently and meet the inspection requirement.