- Posted by iqonic
- On December 16, 2021
- 0 Comments
- @defect, #semiconductor, #WaferCaps, Manufacturing
On December 10th, 2021 our partner Brunel University London published its second iQonic funded publication titled “An Improved Capsule Network (WaferCaps) for Wafer Bin Map Classification Based on DCGAN Data Upsampling” on the IEEE Transactions on Semiconductor Manufacturing website in open access.
Wafer bin maps contain vital information that helps semiconductor manufacturers to identify the root causes and defect pattern failures in wafers. Automatic inspection techniques are studied to sustitue manual inspection techniques labour intensive and cause of prolonged production cycle time.
Brunel proposed WaferCaps architecture
This paper proposes a deep learning approach based on deep convolutional generative adversarial network (DCGAN) and a new Capsule Network (WaferCaps) to increase and classify the samples of the well-known WM-811K dataset of semiconductor wafer defect patterns. The proposed approach has shown an effective performance in generating new synthetic data and classify them with training accuracy of 99.59%, validation accuracy of 97.53% and test accuracy of 91.4%.
The article will appear in the IEEE Transactions on Semiconductor Manufacturing Issue 1, volume 35, year 2022 available in Gold open Access.