A Multiclass Classification Method Based on Convolutional Neural Network for Leaf Identification

The inspection of leaf categorization system is a very important subject in the field of Botany. A classifier of leaf with high accuracy will also bring a lot of fun to people's lives. However, because of the complex background of leaves, the similarity between the different species of leaves, and the differrnces among the same species of leaves, there are still some advance in the recognition of leaf images and their exposer to various diseases. Due to this complication it is often difficult to use the appropriate Pesticides. Based on Inception-v3 model of TensorFlow platform, a methodology is projected to identify the leaf and classify the type of diseases that the leaf has been affected by, using the available dataset a neural network is trained to meet the required application and detect the type of disease and displays the possible medicines.