Logo image
Improving CNN-based Pad Defect Classification with Enhanced Preprocessing Techniques
會議論文集

Improving CNN-based Pad Defect Classification with Enhanced Preprocessing Techniques

Chia-Feng Chien, Ming-Wei Chen, Yin-Chen Chang, Yung-Chih Chen, Yan-Hsiu Liu, Hao-Chiang Shao, Chia-Wen Lin, Yi-Ting Li, Wuqian Tang 和 Chun-Yao Wang
International SoC Design Conference, 頁碼.1-2
15/10/2025
Web of Science ID: WOS:001786518800136

摘要

Accuracy Adaptation models background removal binary classification Companies Convolutional neural networks Feature extraction Interference pad defect Production Semiconductor device manufacture Semiconductor device modeling Noise
In the domain of binary classification, Convolutional Neural Networks (CNNs) are a well-established technique, evidenced by numerous studies highlighting their superior accuracy and generalizability compared to other methods. We try to use CNNs on a semiconductor manufacturing company's pad defect dataset to evaluate their effectiveness in classification. However, due to the significant influence of noise, the results are unsatisfactory. Thus, we propose a preprocessing technique, which includes U-Net pad feature extraction and dilation operation in this work. This preprocessing technique allows the model to identify the features of pad defects more accurately, minimizing the impact of background interference. Consequently, the proposed approach is more adaptable across various production lines in that company.

相關連結

指標

1 檢視次數

詳細資料

Logo image