“Python Code for Accessing Numina and Analyzing Data.”-Using Computer Vision Data to Evaluate Bicycle and Pedestrian Improvements [supporting dataset]

Abstract of the final report is stated below for reference: The design of urban streets and sidewalks must balance the mobility and safety requirements of pedestrians, cyclists, transit users, freight vehicles, and auto drivers. Modern engineers and designers supplement traditional traffic engineering approaches that focus on metrics for automobiles and trucks with more formal consideration of people traveling by foot, bicycle, and bus. The advent of computer vision systems that record counts and pathways of all street and sidewalk users has created new opportunities for data collection, supporting insights that are not possible from the pneumatic tube counters or electromagnetic sensors that are commonly used to measure car and truck traffic. The resulting data can inform street designs that better accommodate all travelers and modes with greater safety. However, computer vision systems using emerging technology must demonstrate their ability to generate consistent, valid, and actionable data before their widespread adoption by urban designers and engineers.

• Dataset URL:
  • https://doi.org/10.7910/DVN/QIPCPS
• Record URL:
  • https://hdl.handle.net/1773/52662
• Supplemental Notes:
  • The dataset supports report: Using Computer Vision Devices to Evaluate Bicycle and Pedestrian Improvements, available at the URL above. This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.
• Corporate Authors:

  University of Washington, Seattle

  Seattle, WA  United States  98195

  University of Idaho

  ,    

  Pacific Northwest Transportation Consortium

  University of Washington
  More Hall Room 112
  Seattle, WA  United States  98195-2700

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Authors:
  • Lowry, Michael
  • 0000-0001-8485-4799
  • MacKenzie, Don
  • 0000-0002-0344-2344
• Publication Date: 2025-1-3

Language

• English

Media Info

• Media Type: Dataset
• Dataset: Version: 1.0 Integrity Hash:
• Dataset publisher:

  Harvard Dataverse

  ,    

Subject/Index Terms

• TRT Terms: Bicycle lanes; Data; Machine vision; Pedestrian traffic; Sensors; Sidewalks; Traffic safety; Urban design
• Identifier Terms: Python (Programming Language)
• Geographic Terms: Boise (Idaho)
• Subject Areas: Data and Information Technology; Design; Highways; Pedestrians and Bicyclists; Safety and Human Factors;

Filing Info

• Accession Number: 01946132
• Record Type: Publication
• Contract Numbers: 69A3551747110
• Files: UTC, NTL, TRIS, USDOT
• Created Date: Feb 18 2025 10:45AM