Automated visual inspection of CMS HGCAL silicon sensor surface using an ensemble of a deep convolutional autoencoder and classifier
This work addresses a domain-specific quality control bottleneck in particle physics experiments, offering an incremental improvement through automation.
The paper tackled the problem of automating visual inspection for anomalies on silicon sensor surfaces in the CMS HGCAL at CERN, achieving an 85% reduction in images needing human inspection with 97% recall.
More than a thousand 8" silicon sensors will be visually inspected to look for anomalies on their surface during the quality control preceding assembly into the High-Granularity Calorimeter for the CMS experiment at CERN. A deep learning-based algorithm that pre-selects potentially anomalous images of the sensor surface in real time has been developed to automate the visual inspection. The anomaly detection is done by an ensemble of independent deep convolutional neural networks: an autoencoder and a classifier. The performance is evaluated on images acquired in production. The pre-selection reduces the number of images requiring human inspection by 85%, with recall of 97%. Data gathered in production can be used for continuous learning to improve the accuracy incrementally.