Infrastructure Camera Video Data Processing of Traffic at Roundabouts
- Delivery
- Provide download link
- Format
- Price
- Non-members (tax incl.)：￥6,600 Members (tax incl.)：￥5,280
- Paper/Info type
- SAE Paper
No.2021-01-0165
- Pages
- 1-14(Total 14 p)
- Date of publication
- Apr 2021
- Publisher
- SAE International
- Language
- English
- Event
- SAE WCX Digital Summit 2021
Detailed Information
| Author(E) | 1) Jagruti Patil, 2) Punit Tulpule |
|---|---|
| Affiliation(E) | 1) The Ohio State University, 2) The Ohio State University |
| Abstract(E) | Roundabout is a unique approach of managing traffic at intersections because it relies on driver’s instincts of safety. Roundabouts are considered safer than other ways of intersection traffic management due to low speed limits, smoother merging, and reduced fatal accidents. Despite their benefits and increasing usage, there is lack of clear understanding of the roundabouts, particularly due to scarcity of data and simulation models and the complexity of the structure. Real-time and offline traffic data recorded at a roundabout provides a basis for 1) identification of the safety issues, 2) understanding unexpected and risky driver behavior, 3) proposing potential mobility solutions, and 4) developing simulation models. The processed data may be used in controlling metered roundabouts, communicating with connected and automated vehicles (CAVs) etc. In this paper an approach to obtain useful traffic information from video feed data at a roundabout is presented. At its core, the proposed approach uses You Only Look Once (YOLOv3) algorithm custom trained for the videos obtained from multiple infrastructure cameras at a roundabout. Roundabout video data processing is challenging because the entire roundabout located in Dublin, OH is captured using eight cameras installed at the roundabout instead of a single camera recording depicting the Bird’s eye view of the roundabout. In this paper we present a way to fuse multiple camera recordings and compute roundabout specific traffic data like - 1) Origin Destination (OD) matrix for all road units defined in terms of road and lane, 2) trajectories of vehicles along the fastest vehicle path (tangential path to the roundabout) and 3) traffic flow rates along different lanes. Some simple diagnostic algorithms are presented to validate the proposed methods. |
