Sidewalk & Curb Audit

Most cities lack systematic sidewalk condition data. Cyvl imagery search can fill the gap by detecting damage and accessibility features in street-level photos.

Workflow

1. Check what official data exists: Search the open data portal for sidewalk datasets: search_datasets("sidewalk")

   • Assess what's available — many cities have geometry (location) data but no condition data.

2. Search imagery for sidewalk conditions — adjust queries to the user's focus:

   • search_imagery("cracked sidewalks and damaged curbs") — general condition
   • search_imagery("ADA wheelchair ramps at intersections") — accessibility inventory
   • search_imagery("wheelchair accessible ramp with tactile paving") — compliant ramps
   • search_imagery("narrow sidewalk blocked by utility pole") — obstructions
   • search_imagery("missing curb ramps") — accessibility gaps

3. If a specific neighborhood is requested, use geocode_location first, then add spatial filters (lat, lon, radius_m).

4. Cross-reference with 311 complaints: Query 311 for sidewalk-related complaints:

   • Legacy 311 (1a0b420d-99f1-4887-9851-990b2a5a6e17): filter case_title ILIKE '%sidewalk%'
   • Returns: 'Sidewalk Repair (Make Safe)', 'Unshoveled Sidewalk'
   • For pothole complaints: case_title ILIKE '%pothole%'
   • Always use execute_sql — query_data filters are broken

5. Present findings:

   • Damaged sidewalks found in imagery (count and confidence)
   • ADA ramp inventory from imagery
   • Accessibility obstructions identified
   • 311 complaint volume for sidewalk issues
   • Gap analysis: how do citizen-reported issues compare to what imagery reveals?

Key Insight

Building a sidewalk inventory manually costs millions. Imagery search generates a preliminary condition inventory from existing street-level photos — today. It can also validate individual repair requests against visual evidence.