#ZeroCrash Project Update Major Progress in the ZeroCrash System
We’re excited to share the latest updates on the ZeroCrash Project, our AI-powered system for predicting and preventing road crashes before they happen.
# What’s New
Streamlit Dashboard: Integrated a complete web interface where users can upload CCTV images, fetch weather data, and enter driver reports in one place.
Model Integration: Successfully combined three independent models — Vision (Vertex AI), Weather (XGBoost), and Driver Text (Gemini API) — into a unified prediction system.
Weather Prediction Upgrade: The weather-based crash prediction model now uses XGBoost, showing improved accuracy and stability compared to previous baselines.
Text Classification: Gemini API now classifies driver feedback in real time to detect risky road conditions from text or voice reports.
Improved Workflow: Streamlined data flow between the weather model, vision model, and text model for faster combined predictions.
Tech Stack Snapshot
Python | Streamlit | XGBoost | Scikit-learn | TensorFlow | Google Cloud Vertex AI | Gemini API | Weather API
Next Steps
Automating real-time CCTV video input directly from traffic feeds.
Integrating live weather API updates.
Enhancing dashboard visualization with dynamic charts.
Launching the public demo link via Streamlit Cloud or Google Cloud Run.D