Plant Care API

A Flask-based API that uses OpenAI's models to provide plant diagnosis and chat functionality.

Features

• Plant Diagnosis: Upload plant images for AI-powered health diagnosis
• Chat Support: Ask questions about plant care and get AI responses
• Cost-Effective: Uses the most affordable OpenAI models for each use case
• Multiple Model Support: Automatic fallback between models for reliability

Setup

1. Install Dependencies

   pip install -r requirements.txt

2. Set up Environment Variables

   • Copy .env.example to .env
   • Add your OpenAI API key:

   OPENAI_API_KEY=your_openai_api_key_here
   

3. Run the Application

   python main.py

The API will be available at http://localhost:5000

API Endpoints

1. Plant Diagnosis - /plant/diagnose (POST)

Upload a plant image for AI-powered diagnosis.

Request:

• Method: POST
• Content-Type: multipart/form-data
• Fields:
  • image (required): Plant image file (PNG, JPG, JPEG, GIF, BMP, WEBP)
  • additional_info (optional): Additional context about the plant
  • label (optional): User-provided plant/crop name

Response:

{
  "success": true,
  "message": "Plant diagnosis completed successfully",
  "diagnosis": {
    "name": "Tomato",
    "status": "nutrient_deficient",
    "confidence": 85,
    "problem": "Yellowing leaves on lower branches with brown edges",
    "cause": "Nitrogen deficiency due to poor soil nutrition",
    "treatment": "Apply balanced fertilizer every 2 weeks and ensure proper watering",
    "prevention": "Maintain consistent watering schedule and soil pH between 6.0-6.8"
  },
  "model_used": "gpt-4o-mini",
  "additional_info": "User provided context",
  "user_label": "Tomato Plant"
}

2. Chat - /chat (POST)

Ask questions about plant care and get AI responses.

Request:

• Method: POST
• Content-Type: application/json
• Body:

{
  "message": "How often should I water my succulents?",
  "conversation_history": [
    {"role": "user", "content": "Previous question"},
    {"role": "assistant", "content": "Previous answer"}
  ]
}

Response:

{
  "success": true,
  "message": "Chat response generated successfully",
  "response": "AI response to your question",
  "conversation_history": [
    {"role": "user", "content": "Your question"},
    {"role": "assistant", "content": "AI response"}
  ],
  "model_used": "gpt-3.5-turbo"
}

Model Configuration

The API uses different OpenAI models optimized for cost and performance:

• Diagnosis: gpt-4o-mini (most affordable vision model)
• Chat: gpt-3.5-turbo (most affordable for conversations)

Fallback models are automatically used if the primary models fail.

Status Values

The diagnosis endpoint returns health status with these possible values:

• healthy: Plant appears to be in good condition
• unhealthy: Plant shows signs of poor health
• diseased: Plant has visible disease symptoms
• pest_infested: Plant is affected by pests
• nutrient_deficient: Plant shows nutrient deficiency signs
• stressed: Plant is under stress (environmental, water, etc.)
• unknown: Unable to determine health status

Error Handling

The API includes comprehensive error handling for:

• Invalid file types
• Missing API keys
• OpenAI API errors
• File upload issues
• Malformed requests

Example Usage

Using curl for diagnosis:

curl -X POST http://localhost:5000/plant/diagnose \
  -F "image=@path/to/plant_image.jpg" \
  -F "label=Tomato Plant" \
  -F "additional_info=My plant has been wilting for 3 days"

Using curl for chat:

curl -X POST http://localhost:5000/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "What are the signs of overwatering in houseplants?"}'

Development

The project follows a modular structure:

• services/openai_client.py: OpenAI integration and model management
• routes/diagnose/: Plant diagnosis endpoint
• routes/chat/: Chat endpoint
• main.py: Flask application setup

License

This project is for educational and development purposes.