Inspection Methodology for Improving Wafer Quality in Semiconductor Production

The edge grinding process refers to the process of polishing the edge of the wafer during the wafer manufacturing process. Visual inspection and rule-based vision inspection are used to detect defective conditions after the process. In the case of visual inspection, the detection rate of defects are different depending on the skill level of the operator, so the defect classification results may be inconsistent. When the rule-based method is used, defects may not be detected or may be over-detected. To solve this problem, this paper is proposed a method to develop an automated system for the grinding process through vision inspection using deep learning. The proposed method is expected to strengthen responses to low productivity and quality problems and improve the competitiveness of the semiconductor industry. Recently, the demand for automation equipment is increasing due to the growth of semiconductor wafer factories, and the methodology proposed in this paper is expected to make a positive contribution to this trend.