Oversampling Based on Data Augmentation in Convolutional Neural Network for Silicon Wafer Defect Classification

Batool, Uzma; Shapiai, Mohd Ibrahim; Ismail, Nordinah; Fauzi, Hilman; Salleh, Syahrizal

doi:10.3233/faia200547

Cited by 3 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several studies have demonstrated the effectiveness of deep learning for silicon wafer defects [26,27] with a focus on single-type defects [28,29,30,31,32]. However, there is a need to investigate more mixed-type defects, which are becoming common with the increasing complexity of fabrication processes.…”

Section: Defect Segmentationmentioning

confidence: 99%

An Attention-Augmented Convolutional Neural Network With Focal Loss for Mixed-Type Wafer Defect Classification

Batool,

Shapiai,

Mostafa

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Silicon wafer defect classification is crucial in improving fabrication and chip production. While deep learning methods have been successful in single-defect wafer classification, the increasing complexity of the fabrication process has introduced the challenge of multiple defects on wafers, which requires more robust feature learning and classification techniques. Attention mechanisms have been used to enhance feature learning for multiple wafer defects. However, they have limited use in a few mixed-type defect categories, and their performance declines as the number of mixed patterns increases. This work proposes an attention-augmented convolutional neural networks (A2CNN) model for enhanced discriminative feature learning of complex defects. The A2CNN model emphasizes the features in the channel and spatial dimensions. Additionally, the model adopts the focal loss function to reduce misclassification and a global average pooling layer to enhance the network's generalization by reducing overfitting. The A2CNN model is evaluated on the MixedWM38 wafer defect dataset using 10-fold cross-validation. It achieves impressive results, with accuracy, precision, recall, and F1-score reported as 98.66%, 99.0%, 98.55%, and 98.82% respectively. Compared to existing works, the A2CNN model performs better by effectively learning valuable information for complex mixed-type wafer defects.

show abstract

Section: Defect Segmentationmentioning

confidence: 99%

An Attention-Augmented Convolutional Neural Network With Focal Loss for Mixed-Type Wafer Defect Classification

Batool,

Shapiai,

Mostafa

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In this paper, geometric transformations are used on the oversampled data. One of the following geometric transformations is randomly applied on each image, rotation by (0, 90, 180, 270) and flipping horizontally or vertically [34].…”

Section: Data Preprocessingmentioning

confidence: 99%

Classification of Human Protein in Multiple Cells Microscopy Images Using燙NN

Al-joudi,

Arif

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

The subcellular localization of human proteins is vital for understanding the structure of human cells. Proteins play a significant role within human cells, as many different groups of proteins are located in a specific location to perform a particular function. Understanding these functions will help in discovering many diseases and developing their treatments. The importance of imaging analysis techniques, specifically in proteomics research, is becoming more prevalent. Despite recent advances in deep learning techniques for analyzing microscopy images, classification models have faced critical challenges in achieving high performance. Most protein subcellular images have a significant class imbalance. We use oversampling and under sampling techniques in this research to overcome this issue. We have used a Convolutional Neural Network (CNN) model called GapNet-PL for the multi-label classification task on the Human Protein Atlas Classification (HPA) Dataset. Authors have found that the Parametric Rectified Linear Unit (PreLU) activation function is better than the Scaled Exponential Linear Unit (SeLU) activation function in the GapNet-PL model in most classification metrics. The results showed that the GapNet-PL model with the PReLU activation function achieved an area under the ROC curve (AUC) equal to 0.896, an F1 score of 0.541, and a recall of 0.473.

show abstract

“…To increase the quantity and diversity of the scarce training data, Kang [61] demonstrated the effectiveness of rotation-based data augmentation in wafer map pattern classification, adopting a LeNet5 like CNN. Batool et al [62] performed augmentation by rotating and flipping images for imbalance management and robust training.…”

Section: A: Custom-made Cnn For Single-label Defect Classificationmentioning

confidence: 99%

A Systematic Review of Deep Learning for Silicon Wafer Defect Recognition

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

Oversampling Based on Data Augmentation in Convolutional Neural Network for Silicon Wafer Defect Classification

Cited by 3 publications

References 0 publications

An Attention-Augmented Convolutional Neural Network With Focal Loss for Mixed-Type Wafer Defect Classification

An Attention-Augmented Convolutional Neural Network With Focal Loss for Mixed-Type Wafer Defect Classification

Classification of Human Protein in Multiple Cells Microscopy Images Using燙NN

A Systematic Review of Deep Learning for Silicon Wafer Defect Recognition

Contact Info

Product

Resources

About