Plant-Disease-Detection-and-Identification

This model learns all the features of 48 different kinds of plants (Healthy and diseased) from the PlantVillage Dataset, and identifies the type of disease and the plant when you input any image in the model.

Uses 5 layered Conv2D Network, and is trained on 10 epochs as of now. It gives an accuracy of 86% on the valdiation set.

Note :

This project is also being extended, wherein we can identify the cures of the diseases and display them along with the disease. Long-term goal is to deploy this model to run offline on mobile-phones.

For visualizing the outputs, please visit the Kaggle link given below :

https://www.kaggle.com/theskrill/plant-disease-detection-plantvillage/output?scriptVersionId=36117223

P.S. : Couldn't add the .h5 and the .pkl file because the file size was too large for GitHub.