Convolution Neural Network Model for Image Classification of Graphene Oxide Thin Films Sheet Resistance

The microscopic images indicate the presence of few layers of graphene sheet resistance with a morphology resembling a thin curtain, and it is hard to recognize manually without artificial intelligence. Therefore, the purpose of this project is to classify the Graphene Oxide (GO) thin films based on sheet resistance values through microscopic images by using a convolution neural network (CNN) model. The data was split into 80% for training and 20% for testing and validation. There were two datasets used for the classification; GO deposited from Atomizer 2 and GO deposited from Atomizer 3. The data were classified into three classifications based on their sheet resistance value using MATLAB programming with a deep learning toolbox model and pre-trained network Vgg-16. The proposed method has shown the efficiency of a deep learning convolution neural network in classifying the microscopic images of GO through sheet resistance value. The experimental results show that microscopic images of GO from Atomizer 2 has 72.59% validation accuracy while for microscopy images of GO from Atomizer 3 has SS.72% validation of accuracy.