Waveguide quality inspection in quantum cascade lasers: A capsule neural network approach | Brunel University London Publication
- Posted by iqonic
- On September 15, 2022
- 0 Comments
Thanks to the iQonic project our partners Brunel University London published a new paper this month titled ” Waveguide quality inspection in quantum cascade lasers: A capsule neural network approach” in the Expert Systems With Applications journal.
The open access article deals with the growing demand for consumer electronic devices and telecommunications expected to drive the quantum cascade laser (QCL) market. The increase in the production rate of QCLs increases the likelihood of production failures and anomalies. Therefore, it proposes a decision fusion approach using CNN and WaferCaps to classify waveguide defects of quantum cascade lasers (QCLs) in optoelectronic wafers. The proposed decision fusion approach resulted in a better accuracy of using standalone models.
This research provides a framework for early and automatic detection of waveguide QCL defects using deep learning and computer vision techniques, highly relevant to WP6 of the iQonic project activities concerning an early malfunction detection and prediction inference engine implemented on ALPES use case data-logs, deriving single station near real-time defect assessment models. The respective results were interfaced with MES and other higher level management systems.
Figure 1 –Waveguide quality inspection using a decision fusion of WaferCaps and CNN.
Read the original article here https://www.sciencedirect.com/science/article/pii/S0957417422015238.