Browsing by Subject "Vehicle classification"

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Bayesian Methods for Estimating Average Vehicle Classification Volumes
    (Minnesota Department of Transportation, 1999-10) Davis, Gary A.; Yang, Shimin
    This report describes the development of a data-driven methodology for estimating the mean daily traffic (MDT) for different vehicle classes from short classification-count samples. Implementation of the methodology requires that an agency maintain a small number of permanent classification counters (PCC), whose output is used to estimate parameters describing their monthly and day-of-week variation patterns and covariance characteristics. The probability of a match between a short classification count sample and each of the PCCs is computed, as well as the estimates of the short-counts site's MDTs which would arise if the short-count site had variation patterns identical to each of the PCCs. The final MDT estimates are then simply the weighted averages of these component MDTs, with the matching probabilities providing the weights. Empirical evaluation of the methods using data collected at the Long Term Pavement Performance Project sites in Minnesota indicated that a reliable match of a short-count site could be made using a sample consisting of a one-day classification count from each month of the year. An evaluation of two-day classification count samples indicated that a two-day count is not sufficient to reliably match the site to a factor group, justifying estimation of MDT using weighted averages. For estimating combination vehicle MDT, these samples should be taken between May and October, and between Tuesday and Thursday. In this case the estimated MDT differed on average by about 10% - 12% compared to estimates based on full year's worth of counts, and differed by less than 26%, 95% of the time.
  • Thumbnail Image
    Item
    Data Mining of Traffic Video Sequences
    (University of Minnesota Center for Transportation Studies, 2009-09) Joshi, Ajay J.; Papanikolopoulos, Nikolaos
    Automatically analyzing video data is extremely important for applications such as monitoring and data collection in transportation scenarios. Machine learning techniques are often employed in order to achieve these goals of mining traffic video to find interesting events. Typically, learning-based methods require significant amount of training data provided via human annotation. For instance, in order to provide training, a user can give the system images of a certain vehicle along with its respective annotation. The system then learns how to identify vehicles in the future - however, such systems usually need large amounts of training data and thereby cumbersome human effort. In this research, we propose a method for active learning in which the system interactively queries the human for annotation on the most informative instances. In this way, learning can be accomplished with lesser user effort without compromising performance. Our system is also efficient computationally, thus being feasible in real data mining tasks for traffic video sequences.
  • Thumbnail Image
    Item
    Development of Data Warehouse and Applications for Continuous Vehicle Class and Weigh-in-Motion Data
    (Minnesota Department of Transportation, 2009-10) Kwon, Taek M.
    Presently, the Office of Transportation Data & Analysis (TDA) at the Minnesota Department of Transportation (Mn/DOT) manages 29 Vehicle Classification (VC) sites and 12 Weigh-in-Motion (WIM) sites installed on various Minnesota roadways. The data is collected 24/7 from all sites, resulting in a large amount of data. The total amount of data is expected to substantially grow with time due to the continuous accumulation of data from the present sites and future expansion of sites. Therefore, there is an urgent need to develop an efficient data management strategy for dealing with the present needs and future growth of this data. The solution proposed in this research project is to develop a centralized data warehouse from which all applications can acquire the data. The objective of this project was to develop software for creating a VC/WIM data warehouse and example applications that utilize it. This project was successfully completed by developing the software necessary to build the VC/WIM data warehouse and the application software packages that utilize the data. The main contribution of this project is that it provides a single access point for querying all of the Mn/DOT’s WIM and VC data, from which many more applications can be developed without concerns of proprietary binary formats.
  • Thumbnail Image
    Item
    Improving Freeway Traffic Speed Estimation Using High-Resolution Loop Detector Data
    (Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota, 2013-04) Liu, Henry; Sun, Jie
    In this project, we developed an innovative methodology to solve a long-standing traffic engineering problem, i.e. measuring traffic speed using data from single inductive loop detectors. Traditionally, traffic speeds are estimated using aggregated detector data with a manually calibrated effective vehicle length. The calibration effort (usually through running probe vehicles), however, is time consuming and costly. Instead of using aggregated data, in this project, our data collection system records every vehicle-detector actuation "event" so that for each vehicle we can identify the time gap and the detector occupation time. With such high-resolution "event-based" data, we devised a method to differentiate regular cars with longer vehicles. The proposed method is based on the observation that longer vehicles will have longer detector occupation time. Therefore, we can identify longer vehicles by detecting the changes of occupation time in a vehicle platoon. The "event-based" detector data can be obtained through the implementation of the SMART-Signal (Systematic Monitoring of Arterial Road Traffic Signals) system, which was developed by the principal investigator and his students in the last five years. The method is tested using the data from Trunk Highway 55, which is a high-speed arterial corridor controlled by coordinated traffic signals. The result shows that the proposed method can correctly identify most of the vehicles passing by inductive loop detectors. The identification of long vehicles will improve the estimation of effective vehicle length on roads. Consequently, speed estimation from the inductive loop detector is improved.
  • Thumbnail Image
    Item
    Intersection Decision Support Surveillance System: Design, Performance and Initial Driver Behavior Quantization
    (Minnesota Department of Transportation, 2007-08) Alexander, Lee; Cheng, Pi-Ming; Donath, Max; Gorjestani, Alec; Menon, Arvind; Shankwitz, Craig
    In rural Minnesota, approximately one-third of all crashes occur at intersections. Analysis of crash statistics and reports of crashes at rural expressway through-stop intersections shows that, for drivers who stop before entering the intersection, the majority of crashes involve an error in selecting a safe gap in traffic. The Intersection Decision Support system, developed at the University of Minnesota, is intended to reduce the number of driver errors by providing better information about oncoming traffic to drivers stopped at intersections. This report deals primarily with the surveillance technology which serves as the foundation upon which the IDS system will be built. Three components of the surveillance system are described in detail in the body of the report: 1) a Mainline Sensor subsystem; 2) a Minor Road Sensor subsystem; 3) a Median Sensor subsystem. These subsystems include radar units, laser-scanning sensors, and infrared cameras, integrated with a vehicle tracking and classification unit that estimates the states of all vehicles approaching the intersection. The design, installation, performance, and reliability of each of these three subsystems are documented in the report. The report concludes with an analysis of driver gap acceptance behavior at an instrumented intersection. Gap selection is examined as a function of time of day, traffic levels, weather conditions, maneuver, and other parameters. Log-normal distributions describe gaps acceptance behavior at rural, unsignalized expressway intersections.
  • Thumbnail Image
    Item
    Investigating inductive loop signature technology for statewide vehicle classification counts
    (Minnesota Department of Transportation, 2018-10) Liao, Chen-Fu
    An inductive loop signature technology was previously developed by a US Department of Transportation (DOT) Small Business Innovation Research (SBIR) program to classify vehicles along a section of the roadway using existing inductive loop detectors installed under the pavement. It was tested and demonstrated in California that the loop signature system could obtain more accurate, reliable and comprehensive traffic performance measures for transportation agencies. Results from the studies in California indicated that inductive loop signature technology was able to re-identify and classify vehicles along a section of roadway and provide reliable performance measures for assessing progress, at the local, State, or national level. This study aimed to take advantage of the outcomes from the loop signature development to validate the performance with ground truth vehicle classification data in the Twin Cities Metropolitan Area (TCMA). Based on the results from individual vehicle class verification, class 2 vehicles had the highest match rate of 90%. Possible causes of classification accuracy for other vehicle classes may include types of loops, sensitivity of inductive loops that generates a shadow loop signal on neighboring lanes, and classification library that was built based on California data. To further understand the causes of loop signature performance and improve the classification accuracy, the author suggests performing additional data verification at a permanent Automatic Traffic Recorder (ATR) site. There is also an opportunity to investigate the classification algorithm and develop an enhanced pattern recognition methodology based on the raw loop signature profile of various types of vehicles in Minnesota.
  • Thumbnail Image
    Item
    Refining Inductive Loop Signature Technology for Statewide Vehicle Classification Counts
    (Minnesota Department of Transportation, 2021-12) Liao, Chen-Fu
    Transportation agencies in the U.S. use devices such as loop detectors, automatic traffic recorders (ATR), or weigh-in-motion (WIM) sensors to monitor the performance of traffic network for planning, forecasting, and traffic operations. With a limited number of ATR and WIM sensors deployed throughout the state roadways, temporary double tubes are often deployed to get axle-based vehicle classification counts. An inductive loop signature technology previously developed by a Small Business Innovation Research (SBIR) program sponsored by the US Department of Transportation is used to classify vehicles using existing loops. This technology has the potential to save time and money while providing the state, counties or cities more data especially in the metro area where loop detectors have already been installed. This research leveraged the outcomes from previous development to validate the classification accuracy with video data. A loop signature system was initially installed at a traffic station in Jordan, MN, to evaluate its performance. The system was later moved to another location on US-52 near Coates, MN, to validate its classification accuracy with more heavy- vehicle traffic. Individual vehicle records were manually verified and validated with ground-truth video data using both the 13 and 7-bin classification schemes from the Federal Highway Administration (FHWA) and the Highway Performance Monitoring System (HPMS). The combined results from both test sites indicated that the loop signature technology had an overall classification accuracy of 93% and 96% using the FHWA and HPMS schemes, respectively. The classification performance can be further improved by including additional vehicle signatures from the state to the classification library.