Extraction of Truck Traffic Data Using Video Based Vehicle Detection (Video Detection Based Truck Traffic Data Collection)

Planning and Design of efficient freight transportation infrastructure is critical for the growth of the state and national economy. Planning and Design of efficient freight transportation facilities and networks should consider current and near-future traffic flows and demands to-and-from seaports, airports and adjacent roadway networks. To facilitate this, accurate and comprehensive traffic data extraction of freight (Truck) data on major roadways at macroscopic and microscopic levels are essential. The current data collection systems employed by Regional Trans. Comm. and the Nevada Department of Transportation (NDOT) around the Las Vegas valley do not meet the current practical needs of advanced traffic management systems. Therefore, this project targets at developing a video based (freight) truck data extraction system to determine traffic flow characteristics like volume, average speed, density and classification of trucks with respect to lanes, time, day, month, etc. The extracted data will be used in computer simulation modeling that can be used to analyze the existing infrastructure to determine where inefficiencies are or where they may occur in the near future. A typical video based vehicle detection system consists of camera, video processing system, and/or a communication module or a storage module. The video based vehicle detector system (VVDS) is configured to collect various vehicle flow characteristics based on the application of the system. The virtual detectors and reference layouts are imaginary lines and boxes that are drawn on the snap shot acquired by the camera. The configuration parameters of the VDS include: 1) Height of the camera, 2) Dimensions and reference lines on the view of the camera, 3) Placement of virtual sensors, 4) Modeling of detector functions which combines the normal outputs of two or more detectors into one customized output, etc. There exist various freight transportation data from local, federal and private industries. However, this data is less coherent and adhoc raw data that is difficult for analysis. The research team will assemble such data into a unified data model for the purpose of identifying locations of video data collection and for validation of extracted freight data.

• 

Language

• English

Project

• Status: Completed
• Funding: $75500.00
• Contract Numbers:

  DTRT06-G-0034

• Sponsor Organizations:

  Research and Innovative Technology Administration

  University Transportation Centers Program
  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590

  Regional Transportation Commission of Southern Nevada

  600 S Grand Central Pkwy
  Suite 350
  Las Vegas, NV  United States  89106
• Performing Organizations:

  Transportation Research Center

  University of Nevada, Las Vegas
  4505 Maryland Parkway, MS 4007
  Las Vegas, NV  United States  89154
• Principal Investigators:

  Dangeti, Mukund

  Muthukumar, Venkatesan

• Start Date: 20081201
• Expected Completion Date: 0
• Actual Completion Date: 20100930
• Source Data: RiP Project 28307

Subject/Index Terms

• TRT Terms: Data collection; Freight traffic; Freight transportation; Infrastructure; Planning and design; Sensors; Traffic data; Truck traffic; Vehicle detectors; Video cameras
• Geographic Terms: Nevada
• Subject Areas: Data and Information Technology; Freight Transportation; Safety and Human Factors;

Filing Info

• Accession Number: 01467993
• Record Type: Research project
• Source Agency: Transportation Research Center
• Contract Numbers: DTRT06-G-0034
• Files: UTC, RIP, USDOT
• Created Date: Jan 3 2013 3:43PM