Automated defect recognition of C-SAM images in IC packaging using

摘要: Ultrasonic C-mode scanning acoustic microscopy (C-SAM) is widely used in the semiconductor industry for reli- ability testing and product inspection due to its non- destructively detect defects IC packaging. However, image interpretation defect identification depend largely on ex- perience of operators, there no recognition system; this partly current systems, which are based computer vision algorithms not robust C-SAM images. A new system appli- cation images described paper. The iconic domain two-dimensional grey-level analysis non-linear Mumford-Shah model used, de- fect achieved through use Support Vector Machines (SVMs). verified experiments a sequence corrupted by synthetically gen- erated noise, bias different shape. remarkable rates indicate that (SVMs) suitable package identification.