晶圓電性測試會產生晶圓圖，用以記錄每顆晶粒之測試結果。若缺陷僅零星且隨機分布，其影響通常較為有限，但若缺陷於特定區域形成群聚、環狀或局部集中等空間分布型態，往往反映製程中存在異常。因此，晶圓缺陷辨識不僅在於判斷缺陷是否存在，更可協助工程師快速篩選可疑晶圓，以理解缺陷的位置、範圍與程度，作為品質管理及製程異常追蹤的重要依據。現有研究多採用深度學習進行晶圓圖分類，雖具有良好的辨識效能，但其決策依據較缺乏可解釋性，且效能評估亦可能受到資料切分影響。
有鑑於此，本研究以公開資料集 WM-811K 為研究對象，把晶圓圖視為具有空間結構的二元格點資料，提出一套特徵少、複雜度低且可解釋的缺陷辨識方法。本研究從缺陷的整體數量、發生位置、空間同質性與局部群聚四個面向萃取七項核心特徵：缺陷覆蓋率、外圈集中程度、Moran’s I、Geary’s C，以及三項SaTScan 群聚特徵，再代入 Logistic Regression、Random Forest 與 XGBoost 等分類模型。成效評估以 by-lot 切分搭配平衡抽樣與 300 次重複模擬，透過 F1 等指標及其分佈衡量效能與穩定性，另以影像紋理特徵與卷積神經網路（CNN）作為比較基準。
結果顯示，在瑕疵、非瑕疵的二分類中，本研究方法僅用七項空間統計特徵就能取得約 0.95 的 F1-score，高於影像紋理特徵基準的 0.89，也在本研究採
之 CNN baseline 下略高於 CNN 的 0.94，且呈現較佳的穩定性。另外，在去除Near-full 的七類缺陷多分類中，本研究方法的 F1-score 約 0.898，在本研究設定下亦高於 CNN baseline 的約 0.79，並保留較直接的特徵解釋性。
整體而言，本研究透過少量空間統計特徵可歸納出晶圓圖中缺陷分布的重要資訊，而且在維持辨識效能的同時，提供直接的影像判讀依據。此方法並非取代深度學習，而是提供一種低維、穩定且易於解釋的替代或輔助途徑。

Wafer electrical testing produces wafer maps that record the test result of every die. Sporadic, randomly scattered defects usually have limited impact, whereas defects forming clusters, rings, or locally concentrated patterns often indicate a process abnormality. Wafer defect recognition is therefore not only about whether defects exist; it also helps engineers screen suspicious wafers quickly and understand their location, extent, and severity, providing an important basis for quality management and process anomaly tracing. Most existing studies adopt deep learning for wafer map classification; such models perform well, but the basis of their decisions is less interpretable, and their reported performance may also be affected by how the data are split.
This study therefore uses the public WM-811K dataset and treats a wafer map as bi-nary lattice data with spatial structure, proposing a defect recognition approach with few features, low complexity, and interpretability. Seven core features are extracted along four aspects of a defect pattern, namely its overall amount, location, spatial homogeneity, and local clustering: defect coverage, concentration in the outer ring, Moran’s I, Geary’s C, and three SaTScan cluster features. They are then fed into Logistic Regression, Random Forest, and XGBoost classifiers. Performance is evaluated under a by-lot split with balanced sampling and 300 repeated simulations, using the F1-score and related metrics and their distributions to assess effectiveness and stability, with image texture features and a convolutional neural network (CNN) as comparison baselines.
The results show that, in binary classification of defective versus non-defective wafers, the proposed approach attains an F1-score of about 0.95 using only seven spatial statistical features, higher than the 0.89 of the image texture baseline and slightly higher than the 0.94 of the CNN baseline adopted in this study, while being more stable. In the seven-class defect classification after removing Near-full, it reaches an F1-score of about 0.898, likewise higher than the approximately 0.79 of the CNN baseline under the settings of this study, while retaining more direct feature interpretability.
Overall, a few spatial statistical features can capture the essential information in a wafer map’s defect distribution and, while maintaining recognition performance, provide a direct basis for visual interpretation. The approach is not meant to replace deep learning, but to offer a low-dimensional, stable, and interpretable alternative or complement.

高崇哲（2025），「強化深偽偵測：以統計方法辨識影像的圖像特徵」，國立政治大學統計學系學位論文。
陳慧霜（2023），「影像分析與深偽影片的偵測」，國立政治大學統計學系學位論文。
黃政嘉（2024），「Deepfake 與 GAN 真偽圖像統計分析」，國立政治大學統計學系學位論文。
Alqudah, R., Al-Mousa, A. A., Abu Hashyeh, Y., & Alzaibaq, O. Z. (2023). A systemic comparison between using augmented data and synthetic data as means of enhancing
wafermap defect classification. Computers in Industry, 145, 103809.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
Chen, T., & Guestrin, C. (2016, August). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794).
Cox, D. R. (1958). The regression analysis of binary sequences. Journal of the Royal Statistical Society: Series B (Methodological), 20(2), 215–232.
Geary, R. C. (1954). The contiguity ratio and statistical mapping. The Incorporated Statis-tician, 5(3), 115–145.
Jeong, I., Lee, S. Y., Park, K., Kim, I., Huh, H., & Lee, S. (2023). Wafer map failure pattern classification using geometric transformation-invariant convolutional neural network. Scientific Reports, 13(1), 8127.
Kang, H., & Kang, S. (2021). A stacking ensemble classifier with handcrafted and con-volutional features for wafer map pattern classification. Computers in Industry, 129,
103450.
Kudrov, M., Bukharov, K., Zakharov, E., Grishin, N., Bazzaev, A., Lozhkina, A., Se-menkin, V., Makhotkin, D., & Krivoshein, N. (2019). Classification of wafer maps de-
fect based on deep learning methods with small amount of data. In 2019 International Conference on Engineering and Telecommunication (EnT) (pp. 1–5). IEEE.
Kulldorff, M. (1997). A spatial scan statistic. Communications in Statistics - Theory and Methods, 26(6), 1481–1496.
Moran, P. A. P. (1950). Notes on continuous stochastic phenomena. Biometrika, 37(1/2), 17–23.
Piao, M., Jin, C. H., Lee, J. Y., & Byun, J.-Y. (2018). Decision tree ensemble-based wafer map failure pattern recognition based on Radon transform-based features. IEEE Trans-
actions on Semiconductor Manufacturing, 31(2), 250–257.
Saqlain, M., Jargalsaikhan, B., & Lee, J. Y. (2019). A voting ensemble classifier for wafer map defect patterns identification in semiconductor manufacturing. IEEE Transactions
on Semiconductor Manufacturing, 32(2), 171–182.
Shin, E., & Yoo, C. D. (2023). Efficient convolutional neural networks for semiconductor wafer bin map classification. Sensors, 23(4), 1926.
Shortridge, A. (2007). Practical limits of Moran’s autocorrelation index for raster class maps. Computers, Environment and Urban Systems, 31(3), 362–371.
Sokal, R. R., & Oden, N. L. (1978). Spatial autocorrelation in biology: 1. Methodology. Biological Journal of the Linnean Society, 10(2), 199–228.
Tsai, T. H., & Wang, C. Y. (2025). Wafer map defect classification using deep learning framework with data augmentation on imbalance datasets. EURASIP Journal on Image
and Video Processing, 2025(1), 6.
Wu, M. J., Jang, J. S. R., & Chen, J. L. (2015). Wafer map failure pattern recognition and similarity ranking for large-scale data sets. IEEE Transactions on Semiconductor
Manufacturing, 28(1), 1–12.
Xu, Q., Yu, N., & Essaf, F. (2022). Improved wafer map inspection using attention mechanism and cosine normalization. Machines, 10(2), 146.
Zheng, H., Sherazi, S. W. A., Son, S. H., & Lee, J. Y. (2021). A deep convolutional neural network-based multi-class image classification for automatic wafer map failure recog-
nition in semiconductor manufacturing. Applied Sciences, 11(20), 9769.