Detecting defects in materials using deep convolutional neural networks
Detecting defects in materials using deep convolutional neural networks
Boyadjian, Quentin and Vanderesse, Nicolas and Toews, Matthew and Bocher, Philippe
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2020
Abstract : This paper proposes representing and detecting manufacturing defects at the micrometre scale using deep convolutional neural networks. The information theoretic notion of entropy is used to quantify the information gain or mutual information of filters throughout the network, where the deepest network layers are generally shown to exhibit the highest mutual information between filter responses and defects, and thus serve as the most discriminative features. Quantitative detection experiments based on the AlexNet architecture investigate a variety of design parameters pertaining to data preprocessing and network architecture, where the optimal architectures achieve an average accuracy of 98.54%. CNNs are relatively easy to perform and give impressive achievements in classification tasks. However, the informational complexity coming from the depth of networks represents a limit to improve their capabilities.