Traffic Monitoring Using Satellite and Ground Data: Preparation for Feasibility Tests and an Operational System

This report documents continued research into the feasibility of using data obtained from satellite images to improve estimates of interest in traffic monitoring programs. Using satellite imagery is attractive for traffic monitoring programs, since imagery would allow wide spatial coverage unobtainable from ground-based sensors. In addition, sensors onboard satellites are off-the-road, and, therefore, there is no disruption to traffic flow or increased hazard to personnel during installation and repair. Moreover, high-resolution satellite imagery will soon be available for the first time in the non-military world. This report addresses the following issues: demonstrating that vehicles can be identified and classified accurately from satellite imagery; developing efficient image processing methods; and determining methods to integrate the imagery with ground-based data and assessing the value of this integration. The authors have been concentrating on the ability of satellite-based data to improve estimates of Average Annual Daily Traffic (AADT) on highway segments and Vehicle Miles Traveled (VMT) over the network of these segments. They describe the methods they developed and coded to simulate traffic patterns and true AADT and VMT statistics and estimate these measures from observations assumed to be obtained from samples of the traffic patterns. The estimation component can use either a traditional-based method (what has traditionally been used to estimate these measures from ground-based sensors) or a model-based method that uses observations more efficiently when the data can be assumed to be compatible with a specified underlying stochastic process. They report the results of numerical studies they conducted using their software. These results indicate the potential of satellite-based data to complement ground-based data and markedly reduce the errors in AADT or VMT estimation or the personnel required to maintain an accuracy level when estimating these parameters.

Language

  • English

Media Info

  • Media Type: Web
  • Edition: Final Report
  • Features: Appendices; Figures; References; Tables;
  • Pagination: 228p

Subject/Index Terms

Filing Info

  • Accession Number: 01155062
  • Record Type: Publication
  • Report/Paper Numbers: FHWA/OH-2000/006
  • Contract Numbers: State Job No. 14658(0)
  • Files: TRIS, ATRI, USDOT, STATEDOT
  • Created Date: Apr 16 2010 1:17PM