Comparative Study of Microscopic Fungii Classification Using Transfer Learning Models

Abstract

Historically, the detection and therapy of fungal disease in humans have relied significantly on direct consultations or assessments conducted by specialized laboratory professionals, specifically mycologists. This research explores t he application of transfer learning models, including MobileNet, Inception, Densenet169, DenseNet121, and EfficientNet, in the classification of fungi images. The primary objective is to determine the model that achieves the highest accuracy in this domain. Through comprehensive experimentation and analysis, it was observed that the DenseNet121 model outperformed the others, achieving an accuracy of 91.01%. This outcome underscores the effectiveness of transfer learning techniques in the context of fungi image classification. The study's findings contribute valuable insights into the development of robust and accurate classification systems for fungal species identification.