RoutesITrust

Personal safety is always a top priority, and with RoutesITrust, you can become more aware of crime in your area and choose the safest option for your next walk, run, or hike.

Comment

Inspiration

RoutesITrust was born from a simple yet powerful idea: what if pedestrians could choose not just the fastest route, but the safest one? We were inspired by the challenges people face when navigating urban environments, especially in cities with varying crime patterns. By combining public crime data with route planning, we envisioned a tool that empowers people to make informed, safer travel decisions.

What it does

RoutesITrust integrates real-world crime incident data with walking route information to provide a safety rating for each potential path. It retrieves multiple routes (e.g., from the Google Maps API) and overlays crime data—measured and weighted by severity—onto these paths. The result is a visual heatmap and risk score for each route, helping users identify the safest way to travel.

How we built it

  • Data Collection: We leveraged publicly available crime data from San Francisco, containing detailed incident information including geographic coordinates and severity scores.
  • API Integration: We used the Google Maps Directions API to fetch alternative walking routes.
  • Geospatial Analysis: Using Python libraries such as Pandas, NumPy, and Folium, we processed the route and crime data. We converted coordinate formats, calculated risk densities, and generated heatmaps to visualize crime concentration along each route.
  • Risk Scoring: We developed an algorithm that aggregates risk scores from nearby incidents along a route and normalizes these against citywide crime density, producing an intuitive safety rating.
  • Web Interface: The project is built with Flask, providing a simple web-based interface where users can input origins and destinations, view route options, and see detailed safety insights.

Challenges we ran into

  • Data Integration: Merging high-precision crime data with route coordinates (which have lower precision) required careful geospatial processing and buffering techniques.
  • Normalization: Determining how to normalize risk scores across varying route lengths and citywide crime patterns was complex, leading us to iterate on our scoring model.
  • API and Deployment Hurdles: Configuring API keys, handling rate limits, and managing different operating system line endings (e.g., LF vs. CRLF) posed technical challenges.
  • Visualization: Creating an intuitive and informative heatmap that accurately reflects risk intensity without overwhelming the user took several iterations of tweaking parameters.

Accomplishments that we're proud of

  • Functional Integration: Successfully combining disparate data sources (crime data and route information) into a coherent and actionable tool.
  • Data-Driven Safety Insights: Our model provides tangible, data-driven insights into route safety, helping users make better decisions.
  • User-Centric Design: Developing a web interface that simplifies complex geospatial analytics into a user-friendly experience.
  • Robust Visualization: Creating detailed heatmaps and risk overlays that effectively communicate spatial crime patterns.

What we learned

  • Geospatial Analysis: We deepened our understanding of geospatial data processing, coordinate transformations, and buffering techniques.
  • API Integration: We gained hands-on experience with integrating external APIs and handling their nuances (like rate limits and key management).
  • Data Normalization & Modeling: Our iterative approach taught us how to effectively normalize and compare data across different scales and domains.
  • Collaboration and Problem-Solving: The challenges we faced improved our troubleshooting, debugging, and cross-functional collaboration skills.

What's next for RoutesITrust

  • Real-Time Data Integration: Incorporate live crime data feeds to provide up-to-date risk assessments.
  • Geographical Expansion: Expand the tool to cover multiple cities and tailor safety metrics to local conditions.
  • Enhanced User Experience: Refine the UI/UX based on user feedback, with features like interactive maps, alternative route suggestions, and personalized safety recommendations.
  • Predictive Analytics: Explore machine learning approaches to forecast crime trends and further enhance route safety scoring.
  • API Integration with Navigation Apps: Work on integrating our safety insights into popular navigation platforms to help users make informed decisions on the go.
Share this project:

Updates