⚠️ Challenges We Faced

🤖 AI Model Verification for Fake/Troll Images

• Classifying false reports easily and verifying reports.
• Ensuring the AI vision model correctly assesses a dog's condition.
• Fine-tuning the model to detect fake reports, with authorities verifying reports on-site.

🔒 Data Privacy & Security

• Protecting user data, especially location-based information and payment details, using Blockchain technology.
• Yet to be implemented – future plans include secure data handling and encryption.

🔄 Integration Issues

• Integrating the RAG model into the adoption page to streamline the adoption process.
• Solutions Implemented:
  • Fixed using Gradio chat interface's API, ensuring smooth communication.

🗺️ Hotspot Map Issues

• Initial Challenges:
  • Missing extra information (dog's image and condition) after sending reports.
  • Map latency and refresh delays.
• Solutions Implemented:
  • Used the Folium Python library for seamless map integration on the website.
  • Integrated images into the Folium map.
  • Included the dog's condition for better visibility and reporting.

📍 Handling GPS Inaccuracy

• Initial Challenge:
  • Some users might submit reports with incorrect or approximate locations, affecting rescue efficiency.
• Solution:
  • Used navigator.geolocation to fetch precise latitude and longitude instead of relying on manual input.
  • Implemented OpenStreetMap’s reverse geocoding API to convert coordinates into a readable address.