Machine learning techniques for nanomaterial applications

Paper ID : 1103-ISCHU

Authors

Neama Sayed Mohammed * Cairo

Abstract

The software development in Nanoscience needs to train numerous scanning electron microscope (SEM) images. However, the price of preparing a large SEM dataset is always very high. The current measurement of nanomaterials features mainly uses manual methods, which will bring about instability errors, especially when measuring a large number of them, we search for a suitable computational method to classify the SEM images on a small dataset. To prepare for identifying the composition of nanowire-fiber-mixtures images, we optimize the performance of image classification between nanowires, fibers, and tips (NFT) due to their geometric similarities. Deep learning provides an efficient, fast way to identify nanomaterial SEM images. Through deep learning methods, the spatial characteristics of nanomaterials can be quickly and accurately measured. The SEM images are analyzed by deep learned techniques where the testing accuracies of 3 convolutional neural network (CNN) models are compared. The ShuffleNet shows the highest accuracies at 97.53%.

Keywords

SEM images, nanomaterials and deep learning

Status: Abstract Accepted (Oral Presentation)