Abstract
On-line optical inspection is become popular, because it can not only improve the product quality but also identify the failure of the process immediately. And hence, not only the production yield can be improved, but also the waste due to failure can be reduced. However, traditional optical inspection machine is limited to detect defects appeared on the surface of the product. In order to detect defects beneath surface, X-Ray image is a powerful tool to apply. Computed tomography can provide the size and shape at any desired section about the defect inside a product while traditional X-Ray image can only provide vague information about the defect. In this dissertation, Simultaneous Algebraic Reconstruction Technique (SART) for computer tomography was adopted to inspect BGA defects. In our simulation study, the effects of various parameters on reconstructed images are also studied. As a result, it was found that the SART with proper parameters used is able to rebuild the 3-D BGA structure properly and the inside flaw can be detected.