WildLearn - Explore Nature with AI

WildLearn is a web-based educational platform designed to help users identify and learn about wildlife (plants and animals). Users can upload images of nature, and the system uses advanced AI to analyze the image, identify the species, and generate an engaging, educational description suitable for a younger audience.

The goal: Empower young people to connect with nature through the magic of AI, fostering a deeper appreciation for the natural world.

Features

• AI-Powered Species Identification: Identify plants and animals from images using Google Gemini.
• Witty & Engaging Descriptions: Get fun, educational descriptions tailored for ~15-year-olds, including scientific names and ecological roles.
• Personal Gallery: Track your discoveries and build a personalized nature journal.
• Featured Species: Explore a curated list of interesting species on the homepage.
• User Accounts: Secure signup and login with email verification to save your history.

System Architecture

The project follows a generic Model-View-Controller (MVC) pattern powered by Flask (Python) on the backend.

1. Frontend (View)

• Technology: HTML5, CSS3, JavaScript.
• Templating: Jinja2 templates (templates/).
• Pages:
  • index.html: Homepage featuring "featured species".
  • upload.html: Image upload interface.
  • my_species.html: Gallery of user's uploaded images.
  • login.html, connect.html, verify_email.html: Authentication flows.

2. Backend (Controller)

• Core: Flask (app.py) handles routing and logic.
• Session Management: Redis via Flask-Session for server-side sessions.
• Authentication: Flask-Login for session management and bcrypt for password hashing.
• Email: smtplib handles verification emails.

3. Database & Storage (Model)

• Database: SQLite (site.db) managed via SQLAlchemy.
  • User Model: Credentials and verification status.
  • Image Model: Links to Supabase URLs and AI-generated descriptions.
• File Storage: Supabase Storage. Images are uploaded to a persistent cloud bucket.

4. AI Engine

• Service: Google Gemini (via google.generativeai).
• Logic: Located in ai_scripts/gen_script.py. The AI first detects if the image contains a species, then generates the educational content.

Directory Structure

• app.py: Main entry point. Configures app, DB, routes.
• ai_scripts/:
  • gen_script.py: Interface with Google Gemini API.
  • supabase_connect.py: Logic for uploading files to Supabase bucket.
  • confirmation_email.py & generate_token.py: Utils for email auth flow.
  • to_dict.py: Helper to parse text files into Python dictionaries.
• auth/:
  • config.py: Environment configuration.
• templates/: HTML templates.
• static/: CSS, JS, and static assets.
• instance/: Instance-specific configuration (ignored by git).
• uploads/: Temporary local storage for uploads.

Detailed Workflows

1. User Authentication

1. Sign Up (/connect): User submits credentials. System creates an unverified account and sends an email.
2. Verification (/confirm/<token>): User clicks link. Account is marked verified.
3. Login (/login): Validated users access their profile and gallery.

2. Image Analysis Pipeline

1. Upload: User POSTs an image.
2. Storage: Image is uploaded to Supabase.
3. AI Analysis: Google Gemini checks for species and generates a description.
4. Save: Result is saved to SQLite linked to the user.
5. Display: User sees the result.

Installation

1. Clone the repository:

   git clone https://github.com/your-username/WildLearn.git
   cd WildLearn

2. Install dependencies: Ensure you have redis-server installed on your system.

   # On Ubuntu/Debian
   sudo apt-get update && sudo apt-get install -y redis-server
   
   # Install Python deps
   pip install -r requirements.txt

3. Environment Setup: You will need to set up your environment variables for:

   • Supabase (URL and Key)
   • Google Gemini API Key
   • Email configuration (for sending confirmations)

   (See auth/config.py for required keys)

4. Run the application: Start Redis (if not running as a service) and then the Flask app:

   redis-server --daemonize yes
   python app.py

Roadmap

• Mobile App Development (iOS/Android)
• Gamification (badges, challenges, leaderboards)
• Augmented Reality (AR) Integration
• Citizen Science Integration

Contact

For questions, bug reports, or collaboration inquiries, please contact: Solomon Stead - solomonstead@gmail.com