• 

  Choose image to predict against CNN

• 

  Gemini Insights from image label classification model

🌱 PlantDoctor: AI-Powered Plant Disease Diagnosis

Inspiration

Growing up with various tomato plants, lemon, and avocado trees getting sick, I have a personal connection to the struggles farmers face. Understanding the impact of plant diseases firsthand inspired me to develop a tool that could assist both small-scale gardeners and large agricultural producers.

What It Does

PlantDoctor is an AI-powered plant disease diagnosis tool that classifies plant diseases using a Convolutional Neural Network (CNN). Once classified, the label is passed to Gemini AI, which generates a refined diagnosis and treatment plan. The output is structured into detailed disease information and actionable recommendations to assist farmers in managing plant health effectively.

How We Built It

• CNN Model: The first step was training a deep learning model on plant disease images. After finishing my Advanced ML final, I was eager to apply theoretical knowledge to a practical solution.
  
• ML Pipeline: Built an end-to-end functional pipeline to preprocess images, classify diseases, and pass results to Gemini AI.
  
• Streamlit UI: On day two, I focused on creating an intuitive web interface, making AI-powered plant disease diagnosis accessible to users.
  

Challenges We Ran Into

• Streamlit components proved tricky when handling save states and preserving session data across different user interactions.
  
• Fine-tuning the CNN model to generalize well across multiple plant species required extensive experimentation.
  
• Managing and structuring Gemini AI responses to ensure high-quality, actionable insights.
  

Accomplishments That We're Proud Of

• Developed a fully functional AI-powered plant disease diagnosis tool.
  
• Successfully integrated Gemini AI for automated insights and treatment suggestions.
  
• Created a scalable backend infrastructure using MongoDB Atlas.
  

What We Learned

• How to build and fine-tune a CNN model for image classification.
  
• Practical implementation of Google Gemini AI for structured response generation.
  
• Deploying and managing MongoDB Atlas clusters for efficient data storage and retrieval.
  

What's Next for PlantDoctor

• Deployment: Finalizing and deploying the Streamlit app while optimizing configurations for smoother user experience.
  
• Dataset Expansion: Collecting and refining user queries to improve the model and make the dataset publicly available.
  
• GCP Integration: Running a PySpark job to push processed data into a GCP bucket for further analysis and scalability.
  

Dataset

New Plant Diseases Dataset

Built With

• gemini
• mongodb
• python
• pytorch
• streamlit