2013 Session: 353

• Evaluating Accuracy of New Algorithm for Extracting Vehicle Tracking Data from Videotaping
  Abstract: A methodology for tracking moving vehicles is presented that overcomes many of the practical limitations of current video taping applications many resulting from traffic and site conditions for the road segment being video-taped. The algorithm presented in this paper provides a sound inexpensive procedure for extracting vehicle tracking data with minimum video taping restrictions. This is achieved through a comprehensive filtering of videotaped images, removal of background distortions, reduced impact of image occlusion, identification and construction of blobs from pixel features, and an accurate link to fixed representative reference points inside of the video frame (Ground Control Points or GCP). The tracking algorithm has been applied to a sample of video-taped vehicle trajectories with parallel GPS geo-referenced information to investigated the effect of placement of GCP and video camera angle on error in vehicle tracking.The number of GCP and the deflection angle from the perpendicular camera sightline to the roadway have a significant effect on the accuracy of the detected vehicle trajectories. Slightly higher errors were noted for a small number of GCP. Accuracy in the tracking algorithm is important for the calibration and validation of microscopic traffic simulation models.
  Authors: Guido, Giuseppe Piero; Vitale, Alessandro; Saccomanno, Frank; Astarita, Vittorio; Giofrè, Vincenzo P.
  Authors: Guido, Giuseppe Piero; Vitale, Alessandro; Saccomanno, Frank; Astarita, Vittorio; Giofrè, Vincenzo P.
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-2475
• Automatic Classification of Road Users' Travel Modes in a Mixed-Traffic Roundabout
  Abstract: The objective of this paper is to present and evaluate an automated road-users classification procedure. The classification is based on the analysis of the motion pattern attributes associated with the trajectories of each road-user type; vehicles, pedestrians and cyclists. A novel approach for features selection is proposed where singular spectrum analysis identifies the main harmonics (speed variation) characterizing the movements trajectories. A constraint-based decision procedure is then applied on the selected features to categorize the road-users. Performance evaluation of the proposed classification is presented. Validation of the procedure is undertaken using real world data set collected at a newly designed mixed traffic roundabout in Greater Vancouver, British Columbia. Satisfactory results were demonstrated and evaluated through performance measures with a reported classification accuracy of around 80 percent. The goal of this research is to improve the understanding of road-users behavior in order to enhance the riding condition and provide an efficient and safe commuting environment. The main benefit of this research is to apply classification as a first step in the activity and behaviour recognition of road-users in traffic scenes.
  Authors: Zaki, Mohamed H.; Sayed, Tarek
  Authors: Zaki, Mohamed H.; Sayed, Tarek
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-2894
• Hidden Markov Models for Vehicle Tracking with Bluetooth
  Abstract: Bluetooth is a short range communication protocol. Bluetooth-enabled devices can be detected using road-side equipment, and each detected device reports a unique identifier. These unique identifiers can be used to track vehicles through road networks over time. The focus of this paper is on reconstructing the paths of vehicles through a road network using Bluetooth detection data. A method is proposed that uses Hidden Markov Models, which are a well-known tool for statistical pattern recognition. The proposed method is evaluated on a mixture of real and synthetic Bluetooth data with GPS ground truth, and it outperforms a simple deterministic strategy by a large margin (30%-50%) in this case.
  Authors: Lees-Miller, John D.; Wilson, R. Eddie; Box, Simon
  Authors: Lees-Miller, John D.; Wilson, R. Eddie; Box, Simon
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-3032
• Analysis of Adaptive Data Fusion Algorithm for Urban Network Application
  Abstract: Due to the development in sensing and communication technology, urban traffic data become increasingly available. This provides excellent opportunity for detailed research on urban traffic flow. The challenge is how to make the best use of these newly available data. This paper analyses the use of different type of data for retrieving the underlying traffic pattern. We present and investigate a data fusion algorithm for integrating heterogeneous traffic data in urban networks. The fusion algorithm is developed based upon the adaptive smoothing method (ASM) proposed by Treiber and Helbing. The objective is to produce a more refined picture of urban traffic through processing and integrating data from different sources in urban network. The filtering and fusion algorithm can work with data collected in different spatio-temporal granularity, with different level of accuracy, and from different kinds of sensors. The accuracy of the fusion algorithm is evaluated on a VISSIM microscopic simulation test-bed. This paper contributes to urban traffic analysis and management.
  Authors: Chow, Andy H. F.; Scarinci, Riccardo; Heydecker, Benjamin G.
  Authors: Chow, Andy H. F.; Scarinci, Riccardo; Heydecker, Benjamin G.
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4189
• Using Signature-Based Vehicle Reidentification to Measure Lane-Changing Maneuvers
  Abstract: This paper provides insight to lane change maneuver data by employing a real-time vehicle re-identification and classification system capable of producing individual vehicle matches and classes based on inductive signatures during congested and uncongested conditions. Vehicle re-identification results for a 0.66 mile multilane freeway segment are compared to manually matched vehicle pairs from video data. Examination of lane change probabilities show that re-identification is capable of reproducing lane change maneuvers with minimal error (root mean square error = 0.0162 and correlation coefficient= 0.927). Differences in lane change probability by level-of-service (LOS), vehicle class, and segment type are also examined. Results show that there is variability in lane change probability by LOS and vehicle class. Although other studies have quantified lane change behavior using vehicle re-identification, none has been successful in obtaining measures during congestion and for separate vehicle classes. Not only would the information gathered from this research be useful in calibrating microsimulation models but also could be used as the basis of real-time traffic calming strategies designed to reduce lane changing at the onset of congestion. In addition to an evaluation of merging behavior using re-identification, improvements to the current re-identification methodology based on lane changing to increase correct classification rates are proposed.
  Authors: Regue, Robert; Hernandez, Sarah Vavrik
  Authors: Regue, Robert; Hernandez, Sarah Vavrik
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-3911
• Portable Roadside Sensors for Vehicle Counting and Speed Measurement
  Abstract: This paper focuses on the development of a portable roadside sensor system for measurement of traffic flow rate, vehicle speeds and vehicle classification. The sensor system consists of wireless anisotropic magnetic devices which do not require to be embedded in the roadway – The devices are placed next to the roadway and measure traffic in the immediately adjacent lane. The vehicle detection algorithm is based on thresholds and speed measurement is based on calculation of cross-correlation between longitudinally spaced sensors. Fast computation of cross-correlation is enabled by using frequency domain signal processing techniques. The calculation of vehicle length follows from using a combination of vehicle speed and vehicle occupancy measurements. Rejection of data from vehicles in non-adjacent lanes is done by using model based position analysis of the magnetic field of vehicles. Data is presented from a large number of vehicles on a regular busy urban road in the Twin Cities in Minnesota. Separately, a high accuracy differential GPS system is used to measure vehicle reference speeds to evaluate the accuracy of the speed measurement from the new sensor system. The speed measurement accuracy is shown to be of the order of 2%. The accuracy in vehicle detection using all of the collected urban road data is 100%.
  Authors: Taghvaeeyan, Saber; Rajamani, Rajesh
  Authors: Taghvaeeyan, Saber; Rajamani, Rajesh
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4430
• Sensor Performance in Measuring Vehicle Length
  Abstract: While most vehicle classification currently conducted in the United States is axle-based, some applications could be supplemented or replaced by length-based data. Common length-based methods are more widespread and can be less expensive, including loop detectors and several types of non-loop sensors (both sidefire and in-road sensors). The most frequently deployed data collection method is loop detectors, and most dual-loop installations have the capability of reporting vehicle lengths.This paper explores field and laboratory tests of loop detectors and non-loop sensors for their performance in determining vehicle length and vehicle speed. Field testing was conducted at four different locations in Minnesota and South Dakota. Ten different commercially available sensors were evaluated.The testing results indicated that across a variety of detection technologies, the loop detectors and non-loop sensors generally reported comparable length and speed data. The research also examined different loop configurations, and found that 6-foot x 6-foot loops performed similarly to 6-foot x 8-foot loops, while 6-foot x 6-foot quadrupole loops performed poorly for vehicles with high beds due to their relatively small magnetic field. Loop detector performance was found to not degrade with the variety of lead-in wire lengths that were tested. Laboratory testing conducted with a loop simulator confirmed the field testing and found that loop detector data is generally repeatable.This paper draws from findings of the Loop and Length Based Vehicle Classification pooled fund project [TPF-5(192)] led by the Minnesota Department of Transportation.
  Authors: Minge, Erik; Petersen, Scott
  Authors: Minge, Erik; Petersen, Scott
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4469
• Statistical Analysis of Mobility Impact of Urban Work Zones with Geocoded Lane Closure and Archived Loop Detector Data
  Abstract: Lane closures as a result of freeway work zone constitute 10% of urban congestion and relate to more than 87,000 annual crashes in the US. Researchers have been studying the mobility characteristics of work zones for many years, focusing on speed reduction, queue length, and capacity based on traffic flow data manually processed or collected for a limited number of work zones. With the increased availability of ITS data, especially geo-coded ITS data, new opportunities emerge for studying and evaluating the mobility impact of work zones. In this study, taking advantages of the comprehensive statewide ITS data archived at Traffic Operations and Safety (TOPS) lab, we correlate the detailed work zone data available through the WisLCS system to the 5-min loop detector data archives using the Wisconsin linear reference system STN(State Truck Network)-Link. Two statistical methods, one-sample percentile value test and two-sample Komogorov-Smirnov(K-S) test, are proposed and implemented to compare the speed and flow characteristics between work zone and non work zone conditions. Neither method requires fitting the traffic flow data to specific types of distribution. Using those tools, we further analyzed the mobility characteristics of freeway work zones within the urban area of Milwaukee, WI, USA in 2010. More than 50% of investigated work zones experienced speed drops and about 15-30% also have reduced volumes. Speed drops are more significant within and downstream of the work zones than the upstream of work zones. .
  Authors: Qu, Tao; Jin, Jing; Cheng, Yang; Parker, Steven; Ran, Bin
  Authors: Qu, Tao; Jin, Jing; Cheng, Yang; Parker, Steven; Ran, Bin
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4561
• Axle and Length Based Vehicle Classification Performance
  Abstract: This study evaluates the performance of three freeway, permanent vehicle classification stations against concurrent video based ground truth. All of the stations have dual loop detectors and a piezoelectric sensor in each lane, providing both axle-based and length-based classification. The evaluation is done at the individual, per-vehicle resolution for each vehicle that passed during the study periods (over 18,000 vehicles, uncongested). While the stations exhibited good performance overall (97% correct), the performance for trucks was far worse, e.g., only 60% of the single unit trucks (SUT) were correctly classified. We diagnosed all of the observed errors and some can be fixed quickly while others cannot. Using data from one site, we revise the classifier to solve almost all of the fixable errors and then test the performance at another location.One chronic error found in this research is intrinsic to the vehicle fleet and may be impossible to correct with the existing sensors; namely, the shorter, SUT have a length range and axle-spacing range that overlaps with passenger vehicles (PV). Depending on the calibration, the error may be manifest as SUT counted as PV or vice versa. One should expect such errors at most classification stations. All subsequent uses of the classification data must accommodate this unavoidable blurring error. The blurring also means that one cannot blindly use an axle classification station to calibrate the boundary between PV and SUT for length-based classification stations, otherwise, the unavoidable errors in the axle-based classification will be amplified in the length-based classification scheme.
  Authors: Kim, Seoungbum; Coifman, Benjamin
  Authors: Kim, Seoungbum; Coifman, Benjamin
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-0058
  Practice-Ready: Yes
• Wireless Accelerometer-Based Automatic Vehicle Classification Prototype System
  Abstract: Automatic Vehicle Classification (AVC) systems provide data about vehicle classes that are used for many purposes. This paper describes a prototype axle count and spacing-based AVC system using wireless accelerometers and magnetometers. The accelerometers detect vehicle axles and the magnetometers report vehicle arrivals and departures and estimate speed. The prototype system is installed on I-80 at Pinole, CA and tested under various traffic conditions. Video images and reports from a nearby commercial Weigh-In-Motion (WIM) station provide ground truth to evaluate the performance of the system, including classification, axle spacing and vehicle counts. The results show the prototype AVC system is reliable in classifying vehicles even under congested traffic with 99% accuracy.
  Authors: Ma, Wenteng; Xing, Daniel; McKee, Adam; Bajwa, Ravneet; Flores, Christopher; Fuller, Brian; Varaiya, Pravin P.
  Authors: Ma, Wenteng; Xing, Daniel; McKee, Adam; Bajwa, Ravneet; Flores, Christopher; Fuller, Brian; Varaiya, Pravin P.
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-2261
  Practice-Ready: Yes
• Tablet-Based Traffic Counting Application Designed to Minimize Human Error
  Abstract: Basic traffic counts are one of the key elements for transportation planning and forecasting. As emerging data collection technologies continue to proliferate, the availability of traffic count data will expand by orders of magnitude. However, availability of data in the quality and quantity does not always guarantee data accuracy and it becomes essential to compare the observed data with ground truth data. However, very little research or guidance is available in ensuring the quality of “ground truth” data, with which the count results of automated technologies can be compared. To address this issue in manual count based ground truth data, an Android Tablet-based manual traffic counting application was developed. Unlike other manual count applications, this application allows the data collectors to replay and toggle through the video in supervisory mode to review and correct counts made in the first pass. For system verification, freeway traffic videos around Metro Atlanta were counted and re-counted using the review function of the application. Initial counts and reviewed counts were compared to assess the improvements in count accuracy that result from the review process. The results indicate that there is a notable benefit of performing a review process. The results also suggest that this application has a potential to minimize human errors and provide more accurate “ground truth” traffic count data for use in transportation planning applications and for model verification.
  Authors: Toth, Christopher Stephen; Suh, Wonho; Elango, Vetri Venthan; Sadana, Ramik; Guin, Angshuman; Hunter, Michael P.; Guensler, Randall
  Authors: Toth, Christopher Stephen; Suh, Wonho; Elango, Vetri Venthan; Sadana, Ramik; Guin, Angshuman; Hunter, Michael P.; Guensler, Randall
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-3130
  Practice-Ready: Yes
• Length-Based Vehicle Classification Schemes and Length-Bin Boundaries
  Abstract: Vehicle classification data is an important component of traffic monitoring programs. While most vehicle classification currently conducted in the United States is axle based, some applications could be supplemented or replaced by length-based data. One challenge with collecting axle-based data is the typically higher deployment cost and reliability issues as compared to length-based systems.This paper reports on analyses of alternative length-based vehicle classification (LBVC) schemes and appropriate length-bin boundaries. The primary analyses use data from a set of 13 Long Term Pavement Performance (LTPP) WIM sites, all in rural areas; with additional analyses conducted using data from 11 Michigan DOT WIM sites located in rural and small urban areas. For most states, the recommended LBVC scheme is a four-bin scheme (motorcycles, short, medium, and long), with an optional “very long” bin recommended for use by states in which significant numbers of longer combination vehicles operate.
  Authors: Weinblatt, Herbert; Minge, Erik; Petersen, Scott
  Authors: Weinblatt, Herbert; Minge, Erik; Petersen, Scott
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-2658
  Practice-Ready: Yes
• Application of Multitouch Technology on Mobile Devices for Intersection Performance Measurement
  Abstract: ABSTRACTTraffic engineers and system analysts rely on timely and accurate data acquisition in order to make decisions that directly impact the safety and efficiency of the transportation system. However, data collection for such a purpose has been a complex and expensive task at intersections, and many people often rely on simulation to evaluate operational plans. This research presents the development and testing of an innovative method to collect turning movement and vehicle delay data at an intersection. It utilizes smartphone devices to take advantage of multi-touch technology for movement identification and object tracking. The algorithms are explained in detail. Preliminary tests have been conducted, and the results of this study have strongly supported the proposed method for its further development and potential field applications.
  Authors: Shao, Chun; Yi, Ping
  Authors: Shao, Chun; Yi, Ping
  Year: 2013
  Document Type: Paper
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4967
  Practice-Ready: Yes
• Axle and Length Based Vehicle Classification Performance
  Authors: Kim, Seoungbum
  Authors: Kim, Seoungbum
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-0058
• Tablet-Based Traffic Counting Application Designed to Minimize Human Error
  Authors: Suh, Wonho
  Authors: Suh, Wonho
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-3130
• Using Signature-Based Vehicle Reidentification to Measure Lane-Changing Maneuvers
  Authors: Hernandez, Sarah
  Authors: Hernandez, Sarah
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-3911
• Analysis of Adaptive Data Fusion Algorithm for Urban Network Application
  Authors: Chow, Andy
  Authors: Chow, Andy
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4189
• Sensor Performance in Measuring Vehicle Length
  Authors: Minge, Erik
  Authors: Minge, Erik
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4469
• Statistical Analysis of Mobility Impact of Urban Work Zones with Geocoded Lane Closure and Archived Loop Detector Data
  Authors: Jin, Jing
  Authors: Jin, Jing
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-4561
• Length-Based Vehicle Classification Schemes and Length-Bin Boundaries
  Authors: Weinblatt, Herbert
  Authors: Weinblatt, Herbert
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Data and Information Technology
  Session: 353
  Paper Number: 13-2658