Enhancing Flower Classification with a Deep Learning Approach Using VGG19 and ImageNet

This paper aims to present an analysis on methods and advancements in floral classification techniques. The following research includes several techniques to investigates different approaches that can be used for categorization and picture analysis. Additionally, the main agenda for proposing this paper is to explore hyperparameter tuning and data augmentation techniques to enhance the workability of the model as well as to get to know how to use a reliable and precise image model on large datasets, such as the images of flowers in this case. The VGG19 model serves as the foundation to study architecture, and transfer learning is applied by utilising previously trained weights. Furthermore, the research examines approaches for data augmentation and hyperparameter tuning to improve the robustness and accuracy of the model. The outcome indicates significant improvements in the performance and accuracy, demonstrating the proposed methodology’s efficiency. The report concludes by highlighting drawbacks, which also includes suggestions for further directions and explore potential options to elaborate the scope of floral classification research.