About the Project

The Potato Disease Classification project aims to develop an efficient model using Convolutional Neural Networks (CNN) and TensorFlow to accurately identify and classify diseases affecting potato plants. The project focuses on leveraging deep learning techniques to enhance the early detection of diseases, thereby contributing to more effective crop management and increased agricultural productivity.

Tech Stack

• Building Model: TensorFlow, Keras, CNN
• Dataset: https://www.kaggle.com/datasets/arjuntejaswi/plant-village
• API: FastAPI
• View: ReactJS, HTML, CSS, JavaScript
• Deployment: Google Cloud Platform (for Model), Heroku (for application)

Feature

• Classify Potato Leaf to detect affected potato plants

• Deploying the TF Model (.h5) on GCP

  gcloud auth login
  

  cd gcp
  gcloud functions deploy predict --runtime python38 --trigger-http --memory 512 --project project_id
  

Result

https://www.loom.com/share/4ae0a789a27f4aa79981b776e0981e8a

Further Development Ideas

• Integrating IoT devices for real-time monitoring of potato plants and disease detection
• Developing a mobile application for farmers to receive notifications and insights on potato plant health and disease classification
• Implementing machine learning algorithms on edge devices for offline disease classification in remote farming areas

Reference: https://cloud.google.com/blog/products/ai-machine-learning/how-to-serve-deep-learning-models-using-tensorflow-2-0-with-cloud-functions