# Introduction The plant analysis is a rapidly evolving field of research, and researchers are always looking for new strategies to improve the precision of models. One major obstacle in plant analysis is the variation within the category, which can make it difficult for machine learning models to learn. The new method developed by the team of researchers, known as FGDCC, aims to overcome this obstacle using classification to discover pseudo-labels that represent a measure of similarity between images. The initial results show that the method can significantly improve the efficiency of machine learning models, even when the data is scarce. The source code for the method is available on GitHub.