[17] - CropDoc

Your Toolkit for Growing Healthier Fields

Comment 2

CropDoc: Your AI-Powered Toolkit for Growing Healthier Fields

Team : Eranda Sooriyarachchi

Project Overview:

Inspiration:

This is my first hackathon and I have lived in Iowa, an agricultural state for quite a while. Iowa has two main crops, soybean and corn. I chose to make an image classifier for the common soybean leaf diseases according to a document published by the Iowa State University ( https://crops.extension.iastate.edu/files/article/PM1989.pdf ). I also thought it would be nice to be able to ask questions about the disease, for which a generative AI chabot was made.

CropDoc is an innovative AI-powered solution designed to assist farmers in Iowa in diagnosing and managing crop diseases efficiently. By leveraging computer vision and natural language processing, CropDoc empowers farmers with the ability to identify crop diseases through image recognition and access expert advice through a conversational chatbot. This project aims to address the critical issue of crop health and increase agricultural productivity in Iowa.

Problem Statement:

Farmers in Iowa face numerous challenges related to crop diseases, which can lead to significant crop yield losses if not detected and managed early. Identifying these diseases promptly is essential for preventing widespread outbreaks and optimizing crop yields.

Solution:

CropDoc combines cutting-edge technologies to provide a comprehensive solution:

Crop Disease Diagnosis via Image Recognition:

Farmers can take pictures of their crops using the CropDoc mobile app. The AI system uses computer vision algorithms to analyze the images and identify crop diseases accurately.

Expert Guidance through Chatbot:

CropDoc incorporates a chatbot with a Large Language Model (LLM) that provides real-time responses to farmers' questions about crop diseases. This chatbot utilizes the RAG (Retrieval Augmented Generation) for the upto date information retrieval of the disease and offers appropriate answers to your questions.

Disease Database:

CropDoc is using GPT 3.5 with Retrieval Augmented Generation from Wikipedia.

Technology Stack:

Computer vision for image analysis

Natural language processing (NLP) for chatbot functionality

FastAi

LangChain

Openai GPT 3.5

Gradio for front-end

Huggingface spaces for hosting

Built With

Share this project:

Updates