Plant Species Classification Using Neural Network

Abstract

The task that is done in our project is to classify and identify the species of different plant using neural network. The work is done mainly for removing the problems that arises when classifying different types of plants in their own category. In this research paper, we propose the use of neural networks for plant species classification. Our approach involves the collection and labeling of a large dataset of plant images, which are then used to train and evaluate the performance of neural network model. The neural network model was trained on a variety of plant images, including leaves and fruits to ensure a diverse and representative dataset. To improve the accuracy of our classification, we also employed data augmentation techniques, such as rotation and scaling, to artificially expand our dataset. Our results show that neural network models can accurately classify plant species, with a top accuracy of 90.2%. Additionally, we found that using a combination of different image types, such as leaves and flowers, improved the classification performance compared to using a single type of image. Overall, our study demonstrates the potential of neural networks for plant species classification and highlights the importance of a diverse and representative dataset for achieving high accuracy. This research has practical applications in fields such as agriculture, where accurate plant species identification is necessary for crop management and yield optimization.