Tackling Inter-class Similarity and Intra-class Variance for Microscopic Image-Based Classification

Automatic classification of aquatic microorganisms is based on the morphological features extracted from individual images. The current works their do not consider inter-class similarity and intra-class variance that causes misclassification. We are particularly interested in case where within a class occurs due to discrete visual changes microscopic In this paper, we propose account for it by partitioning classes with high features. Our algorithm automatically decides optimal number sub-classes be created each them as separate training. This way, network learns finer-grained experiments two databases freshwater benthic diatoms marine plankton show our method can outperform state-of-the-art approaches these microorganisms.