01472413

Component

https://doi.org/10.3141/2339-09

http://scholar.google.com...reno&publication_year=2013

Data on pedestrian activity, including volumes, walking speed, and trajectories, are used by transportation agencies and researchers for planning, design, and analysis. Several technologies are available for automatic collection of pedestrian data; however, all have inherent limitations in either functionality or monetary cost. Also, most technologies provide only counts. This paper proposes the use of an inexpensive motion-sensing device, Microsoft Kinect, which can track multiple people in low light conditions and can be combined with existing video-based daytime tracking. The tracking software and speed estimation methodologies are described, and indoor and outdoor studies show the system’s effectiveness at determining pedestrian volumes and walking speeds. The accuracy of speed data is very satisfactory, with correlation of 98% or more for video data validation speeds. The accuracy of pedestrian volume data varies with traffic conditions; however, in low to moderate traffic conditions its performance is acceptable, with an undercounting error near 8%. The applications of the sensor and its complementarity with other sensors are discussed, as these are the first step toward a multisensor system.

01516644

13-3284

English

pp 80–89

2013

Transportation Research Record: Journal of the Transportation Research Board

9780309287098

Print

Figures (7) ; References (15) ; Tables (1)

Data collection; Motion perception; Pedestrian counts; Pedestrian flow; Pedestrian traffic; Sensors; Speed; Traffic counting; Walking

Data and Information Technology; Operations and Traffic Management; Pedestrians and Bicyclists; Planning and Forecasting; I72: Traffic and Transport Planning

TRIS, TRB, ATRI

Feb 5 2013 12:40PM

• Classification of Bicycle Traffic Patterns in Five North American Cities
• Correlation-Based Clustering of Traffic Data for the Mechanistic–Empirical Pavement Design Guide
• Estimating Annual Average Daily Bicyclists: Error and Accuracy
• Evaluation of Axle-Based and Length-Based Vehicle Classification Stations
• Generating Site-Specific Axle Load Factors for the Mechanistic–Empirical Pavement Design Guide
• Hybrid Approach for Clustering Vehicle Classification Data to Support Regional Implementation of the Mechanistic–Empirical Pavement Design Guide
• Impact of Various Trucks on Pavement Design and Analysis: Mechanistic–Empirical Pavement Design Guide Sensitivity Study with Truck Weight Data
• Length-Based Vehicle Classification Schemes and Length Bin Boundaries
• Locating Traffic Sensors on a Highway Network: Models and Algorithms
• Quality Counts for Pedestrians and Bicyclists: Quality Assurance Procedures for Nonmotorized Traffic Count Data
• Sensor Performance in Measuring Vehicle Length
• Tablet-Based Traffic Counting Application Designed to Minimize Human Error
• Use of Reduction-Effectiveness Ratios to Evaluate Reduced Traffic Data Collection Plans