Inventorying Bus Stop Amenities across the United States Using Google Street View Images and Computer Vision

Bus stops play a crucial role as primary points of interaction between the public and transit systems, facilitating access to essential services and opportunities. However, comprehensive data on bus stop amenities are scarce across U.S. cities, hindering efficient resource allocation to improve bus stops and research efforts within the transportation community. To address this critical data gap, this project proposes the development of a national database of bus stop amenities utilizing Google Street View (GSV) images and advanced computer vision algorithms. Specifically, this project aims to automate the detection and cataloging of bus stop amenities such as shelters, seating, signage, lighting, and accessibility features. To achieve this aim, we will start with the development of computer vision algorithms to detect bus shelters in Florida cities. Subsequently, a national-scale model will be trained using data from diverse regions to ensure applicability across U.S. cities. Furthermore, the project will explore the detection of additional amenities beyond shelters, including seating, bicycle parking, crosswalks, signage, and accessibility features. The main product of this project will be a comprehensive national bus stop census database, which will be made publicly accessible through a user-friendly web interface. This database will be a valuable resource for transit agencies to prioritize investments and enhance rider experience. Additionally, it will serve as an essential piece of research and education infrastructure for the transportation community. By democratizing access to bus stop data, this project will foster innovations in transit research and promote equitable transit-oriented communities. This database can open a wide range of research and application opportunities for transit agencies across the United States and beyond. The PIs will develop a web-based interface to allow public access to the database. Also, they use technology transfer funds available from CETOC and potentially seek other funding opportunities to maintain and update this database to reflect bus stop changes. Since Google Street View images are updated every 1 to 3 years (more frequent in dense areas where bus stops are located), together with the computer vision algorithm being developed from the project, maintaining this database up to date will be feasible with relatively low costs (estimated annual cost is $3,000 - $5,000). We expect to produce 2-3 publications from the project (Year 1). We will present the work at international/national conferences such as ACSP and TRB annual meetings. The project will lead to new computer vision algorithms and bus stop amenity detection methods widely applicable across US cities and beyond. In addition, the development of this national database will establish close relationships between CETOC and various transit agencies.

• Record URL:
  https://www.uno.edu/cetoc
• Supplemental Notes:
  • Funding: $207,000 (USDOT) + $103,500 (matching funds) = $310,500

Language

• English

Project

• Status: Active
• Funding: $310500
• Contract Numbers:

  69A3552348337

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Managing Organizations:

  Center for Equitable Transit-Oriented Communities (CETOC)

  University of New Orleans
  New Orleans, LA  United States 
• Project Managers:

  Kline, Robin

  Danton, Bob

• Performing Organizations:

  University of Florida

  207 Grinter Hall
  PO Box 115500
  Gainesville, Florida  United States  32611

  University of New Orleans

  Department of Planning and Urban Studies
  New Orleans, LA  United States  70148

  University of Utah, Salt Lake City

  City & Metropolitan Planning
  201 South Presidents Circle
  Salt Lake City, UT  United States  84112
• Principal Investigators:

  Yan, Xiang

  Tian, Guang

  Ewing, Reid

• Start Date: 20241001
• Expected Completion Date: 20250930
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program

Subject/Index Terms

• TRT Terms: Bus stop shelters; Computer vision; Public transit; Transit oriented development
• Identifier Terms: Google Street View
• Subject Areas: Data and Information Technology; Planning and Forecasting; Public Transportation;

Filing Info

• Accession Number: 01928835
• Record Type: Research project
• Source Agency: Center for Equitable Transit-Oriented Communities (CETOC)
• Contract Numbers: 69A3552348337
• Files: UTC, RIP
• Created Date: Aug 26 2024 2:53PM