错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

The Classification of Semiconductor Wafer Defects by Means of Feature-based Transfer Learning Approach

  • Lim Shi Xuen,
  • Jesnor Arif Mat Jizat,
  • Ismail Mohd Khairuddin,
  • Ahmad Fakhri Ab. Nasir,
  • Mohd Azraai Mohd Razman,
  • Zahari Taha,
  • Edmund Yuen,
  • Anwar P. P. Abdul Majeed

摘要

Wafer defect detection is a critical part of the semiconductor wafer manufacturing industry. Human-based manual defect detection is tedious and often prone to misjudgement. Hence, there is a push towards automatic wafer defect detection in the industry. This paper shall investigate the efficacy of a transfer learning pipeline that utilizes different pre-trained VGG models, i.e., VGG16 and VGG19, that act as a feature extractor coupled with a Random Forest (RF) model in classifying wafer defect images. The hyperparameter of the RF classifier is further fine-tuned through the grid-search technique via a five-fold cross-validation method. It was shown in the study that the optimised VGG16+RF pipeline provided a better classification of the wafer defects.