AUTHOR=Zhao Peng , Li Chen , Rahaman Md Mamunur , Xu Hao , Yang Hechen , Sun Hongzan , Jiang Tao , Grzegorzek Marcin 

TITLE=A Comparative Study of Deep Learning Classification Methods on a Small Environmental Microorganism Image Dataset (EMDS-6): From Convolutional Neural Networks to Visual Transformers

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2022.792166

DOI=10.3389/fmicb.2022.792166

ISSN=1664-302X

ABSTRACT=Environmental Microorganisms (EMs) have a significant impact on human survival and development, which is attracted many researchers’ attention. Among them, EM-based image classification is one of the indispensable directions. In recent years, deep learning has made brilliant achievements in EM image classification. However, image classification of small EM datasets is still not obtained good research results. In addition, the performance of different classification methods differs greatly on small datasets. Therefore, researchers need to spend a lot of time searching for models with good classification performance and suitable for the current equipment working environment. In order to provide researchers with some reliable references, this paper uses 21 deep learning models to test the classification performance of small datasets. Experimental results show that Xception performs best in classification performance and ViT model training consumes the least time. The ShuffleNet-V2 model has the least amount of parameters.