Automated visual inspection of CMS HGCAL silicon sensor surface using an ensemble of a deep convolutional autoencoder and classifier.
Sonja GrönroosMaurizio PieriniNadezda ChernyavskayaPublished in: CoRR (2023)
Keyphrases
- automated visual inspection
- surface defects
- multiple classifiers
- ensemble learning
- deep learning
- classifier ensemble
- training data
- quality control
- final classification
- restricted boltzmann machine
- ensemble classifier
- training set
- si sio
- machine vision
- combining classifiers
- individual classifiers
- learning algorithm
- decision tree classifiers
- silicon dioxide
- ensemble methods
- weak classifiers
- multiple classifier systems
- majority voting
- sensor networks
- defect detection
- laser scanner
- feature selection
- weak learners
- deep belief networks
- gallium arsenide
- support vector
- support vector machine
- unsupervised feature learning
- machine learning
- decision trees
- pruning algorithm
- d objects
- real time
- three dimensional
- training samples
- svm classifier
- sensor data
- ensemble members
- classifier combination
- neural network
- multi sensor
- low cost
- unsupervised learning
- feature space
- computer vision
- point cloud