Identification and classification of exfoliated graphene flakes from microscopy images using a hierarchical deep convolutional neural network

Identification of exfoliated graphene flakes and classification the thickness are important in nanomanufacturing advanced materials devices. This paper presents a deep learning method to automatically identify classify on Si/SiO2 substrates from optical microscope images. The presented framework uses hierarchical convolutional neural network that is capable new images while preserving knowledge previous model was trained used into monolayer, bi-layer, tri-layer, four-to-six-layer, seven-to-ten-layer, bulk categories. Compared with existing machine methods, showed high accuracy efficiency as well robustness background resolution results indicated pixel-wise 99% identifying classifying flakes. research will facilitate scaled-up manufacturing characterization for