Plant-Disease-Classification

This is a DL based project which classifies the disease of Potato leaf using CNN, tf-serving, and FastAPI.

Introduction :

• This is a Machine Learning, Deep Learning, and Web-based project which will help farmers. The agrotech industry is constantly evolving and seeking innovative solutions to ensure efficient and sustainable farming practices. This project aims to address the challenges of crop disease through Machine Learning and Deep Learning algorithms.

  The first step of the project is to build a dataset of crop images, categorizing them based on healthy and diseased states. The images will be preprocessed to improve their quality and prepare them for feeding into the machine learning and deep learning models. Once the dataset is prepared, it will be used to train and validate the model. The model will use convolutional neural networks (CNN) for image classification and prediction.

  After the model is trained, it will be deployed on a web application that allows farmers to upload images of their crops and receive a prediction on whether the crops are healthy or diseased.

Future changes/Improvements:

• Add other plant disease predictions
• Use Transfer Learning Techniques
• Can use the Image Generator method
• Use Cross Validation
• Use TF-serving for model versioning
• Add another feature to suggest the best crops based on the soil type
• Add another feature to recommend fertilizers based on the soil type.

Dataset Link: https://www.kaggle.com/datasets/arjuntejaswi/plant-village

Web App Link: https://plant-disease-classification.streamlit.app/