Wafer map pattern classification by Combining Convolutioal and Handcrafted Features

Wafer map defect pattern classification with Stacking Ensemble of Convolutioal and Handcrafted Features

Proposed by H.Kang and S.Kang

Hyungu Kang, Seokho Kang* (2021), "A stacking ensemble classifier with handcrafted and convolutional features for wafer map pattern classification", Computers in Industry 129: 103450 (https://www.sciencedirect.com/science/article/pii/S0166361521000579?via%3Dihub)

Methodology

Convolutional Neural Network (CNN)

• Input: wafer map
  • resized to 64x64
• Output: predicted score
• Model: CNN (based on VGG16)

Manual Feature Extraction (MFE)

• Input: handcrafted features of wafer map
  • 59-dim
• Output: predicted score
• Model: FNN (2-layer MLP)

Stacking Ensemble

• Input: predicted score of CNN and MFE
• Output: predicted class
• Model: MLR (Multi-response Linear Regression)

Data

• WM811K

  • 811457 wafer maps collected from 46393 lots in real-world fabrication

  • 172950 wafers were labeled by domain experts.

  • 9 defect classes (Center, Donut, Edge-ring, Edge-local, Local, Random, Near-full, Scratch, None)

  • provided by MIR Lab (http://mirlab.org/dataset/public/)

  • .pkl file downloaded from Kaggle dataset (https://www.kaggle.com/qingyi/wm811k-wafer-map)

  • directory: /data/LSWMD.pkl

Dependencies

• Python 3.8
• Pytorch 1.9.1
• Pandas 1.3.2
• Scikit-learn 1.0.2
• OpenCV-python 4.5.3
• Scikit-image 0.18.3

References

• WM-811K(LSWMD). National Taiwan University Department of Computer Science Multimedia Information Retrieval LAB http://mirlab.org/dataSet/public/
• Nakazawa, T., & Kulkarni, D. V. (2018). Wafer map defect pattern classification and image retrieval using convolutional neural network. IEEE Transactions on Semiconductor Manufacturing, 31(2), 309-314.
• Shim, J., Kang, S., & Cho, S. (2020). Active learning of convolutional neural network for cost-effective wafer map pattern classification. IEEE Transactions on Semiconductor Manufacturing, 33(2), 258-266.
• Kang, S. (2020). Rotation-Invariant Wafer Map Pattern Classification With Convolutional Neural Networks. IEEE Access, 8, 170650-170658.