Deep learning approach to the classification of selected fungi and bacteria

Detection of microorganisms is of particular importance to human health and life, and for the industry in general. For this reason, we want this process to be as fast and precise as possible. We also expect that the automation of this activity (detection of microorganisms) can be widely used in various industries. This article is another attempt to the classification of microorganisms (a small group of selected species listed in ISO 17516:2014-11) that uses a deep learning approach with popular deep neural network (DNN) models. Before classification, the microorganisms were stained by two methods: Congo Red and the modified Gram stain. The classification results for the two staining methods are similar, but greater accuracy was obtained for Congo Red stain. The final results show that the popular DNN models (Xception, ResNet-50) perform well with this challenge and also show that it is not always beneficial to use the pre-trained weights of DNN models.