Demonstrating the Viability of Automated Video Analysis for Mid-Block Traffic Data Collection: A Study in the City of Edmonton
Streamlining traffic data collection is a main goal for cities seeking to improve safety and mobility on their roads. Traffic data collection can benefit greatly from the recent advances in automated video analysis. With the ability of accurate measurements and tracking vehicle coordinates, applications to collect reliable traffic flow are practically possible using computer vision. The aim of this paper is threefold. First, to improve the detection and tracking technology through fusion of computer vision approaches. This task is performed by complementary region based detection to the already established featured based tracking. As a result, a better estimation of the road-users position, size and travel information is now possible. Second, to demonstrate the technology in a new application to implement a vision based virtual traffic measurement system. With particular focus on vehicle classification, and the measurement of basic traffic parameters such as counting, speed, headway and gap time measurements. Third, to test and validate the new development on midblock road segments video data collected by the City of Edmonton Office of Traffic Safety (OTS). In demonstrating the application of video analysis for traffic data collection, this paper seeks to underline the potential benefits which transportation agencies can derive from progressively adopting new technologies in performing everyday tasks.
-
Supplemental Notes:
- This paper was sponsored by TRB committee ABJ35 Standing Committee on Highway Traffic Monitoring.
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Zaki, Mohamed H
- Sayed, Tarek
- Ibrahim, Shewkar E
-
Conference:
- Transportation Research Board 96th Annual Meeting
- Location: Washington DC, United States
- Date: 2017-1-8 to 2017-1-12
- Date: 2017
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; References; Tables;
- Pagination: 20p
- Monograph Title: TRB 96th Annual Meeting Compendium of Papers
Subject/Index Terms
- TRT Terms: Data collection; Data fusion; Image analysis; Traffic data; Vehicle classification; Video imaging detectors
- Geographic Terms: Edmonton (Canada)
- Subject Areas: Data and Information Technology; Highways;
Filing Info
- Accession Number: 01622567
- Record Type: Publication
- Report/Paper Numbers: 17-02117
- Files: TRIS, TRB, ATRI
- Created Date: Jan 17 2017 9:47AM