Browsing by Author "Papanikolopoulos, Nikolaos P"

Now showing 1 - 12 of 12

Algorithms for Vehicle Classification
(Minnesota Department of Transportation, 2000-07-01) Gupte, Surendra; Papanikolopoulos, Nikolaos P
This report presents algorithms for vision-based detection and classification of vehicles in modeled at rectangular patches with certain dynamic behavior. The proposed method is based on the establishment of correspondences among blobs and vehicles, as the vehicles move through the image sequence. The system can classify vehicles into two categories, trucks and non-tucks, based on the dimensions of the vehicles. In addition to the category of each vehicle, the system calculates the velocities of the vehicles and generates counts of vehicles in each lane over a user-specified time interval, the total count of each type of vehicle, and the average velocity of each lane during this interval.
Algorithms for Vehicle Classification: Phase II
(2001-11-01) Martin, Robert; Masoud, Osama; Gupte, Surendra; Papanikolopoulos, Nikolaos P
This report summarizes the research behind a real-time system for vehicle detection and classification in images of traffic obtained by a stationary CCD camera. The system models vehicles as rectangular bodies with appropriate dynamic behavior and processes images on three levels: raw image, blob, and vehicle. Correspondence is calculated between the processing levels as the vehicles move through the scene. This report also presents a new calibration algorithm for the camera. Implemented on a dual Pentium PC equipped with a Matrox Genesis C80 video processing board, the system performed detection and classification at a frame rate of 15 frames per second. Detection accuracy approached 95 percent, and classification of those detected vehicles neared 65 percent. The report includes an analysis of scenes from highway traffic to demonstrate this application.
Automatic Passenger Counting in the HOV Lane
(1999-06-01) Pavlidis, Ioannis; Symosek, Peter; Morellas, Vassilios; Fritz, Bernard; Papanikolopoulos, Nikolaos P; Sfarzo, Robert
This research applied wave band and computer vision methods to automatically count vehicle occupants in the High Occupancy Vehicle (HOV) lane at a high level of accuracy. The research showed that use of near-infrared bandwidth offers potential as a method for developing an automatic vehicle occupant counting system. Near-infrared only can produce images when looking through glass, but not metal or heavy clothes, which limits its accuracy in counting children or occupants resting in vehicles. The mid-infrared camera did not produce clear images at highway speeds. The next step involves additional research into a working device that can count vehicle occupants reliably, including analysis of device performance with more types of vehicles, passengers in the back seats, children in car seats, and passengers lying down.
Bicycle Counter
(2000-03-01) Rogers, Scott; Papanikolopoulos, Nikolaos P
This report describes a system for monitoring bicycle activity in sequences of gray scale images from a stationary camera. Applications for such a system include determining the use and congestion of bicycle paths. The output of the system is a count of the number of bicycles detected in the image sequence. The system uses a simple model of two circular objects separated by relatively known distance, with four levels of abstraction: raw images, blobs, edge images, and the bicycle model. The system was implemented on a dual Pentium computer equipped with a Matrox imaging board and achieved a peak performance of eight frames per second. Experimental results based on outdoor scenes show promising results for a variety of weather conditions.
Development of a Tracking-based Monitoring and Data Collection System
(2005-10-01) Veeraraghavan, Harini; Atev, Stefan; Masoud, Osama; Miller, Grant; Papanikolopoulos, Nikolaos P
This report outlines a series of vision-based algorithms for data collection at traffic intersections. We have purposed an algorithm for obtaining sound spatial resolution, minimizing occlusions through an optimization-based camera-placement algorithm. A camera calibration algorithm, along with the camera calibration guided user interface tool, is introduced. Finally, a computationally simple data collection system using a multiple cue-based tracker is also presented. Extensive experimental analysis of the system was performed using three different traffic intersections. This report also presents solutions to the problem of reliable target detection and tracking in unconstrained outdoor environments as they pertain to vision-based data collection at traffic intersections.
Finding What the Driver Does
(2005-05-01) Veeraraghavan, Harini; Atev, Stefan; Bird, Nathaniel; Schrater, Paul; Papanikolopoulos, Nikolaos P
Most research depends on detection of driver alertness through monitoring the eyes, face, head or facial expression. This research presents methods for recognizing and summarizing the activities of drivers using the appearance of the driver's position, and changes in position, as fundamental cues, based on the assumption that periods of safe driving are periods of limited motion in the driver's body. The system uses a side-mounted camera and utilizes silhouettes obtained from skin color segmentation for detecting activities. The unsupervised method uses agglomerative clustering to represent driver activity throughout a sequence, while the supervised learning method uses a Bayesian eigen image classifier to distinguish between activities. The results validate the advantages of using driver appearance obtained from skin color segmentation for classification and clustering purposes. Advantages include increased robustness to illumination variations and elimination of the need for tracking and pose determination.
Managing Suburban Intersections through Sensing
(2002-12-01) Veeraraghavan, Harini; Masoud, Osama; Papanikolopoulos, Nikolaos P
Increased urban sprawl and increased vehicular traffic have resulted in an increased number of traffic fatalities, the majority of which occur near intersections. According to the National Highway Safety Administration, one out of eight fatalities occurring at intersections is a pedestrian. An intelligent, real-time system capable of predicting situations leading to accidents or near misses will be very useful to improve the safety of pedestrians as well as vehicles. This project investigates the prediction of such situations using current traffic conditions and computer vision techniques. An intelligent system may gather and analyze such data in a scene (e.g., vehicle and pedestrian positions, trajectories, velocities, etc.) and provide necessary warnings. The current work focuses on the monitoring aspect of the project. Certain solutions are proposed and issues with the current implementation are highlighted. The cost of the proposed system is low and certain operational characteristics are presented.
Monitoring Driver Activities
(2004-09-01) Wahlstrom, Eric; Masoud, Osama; Papanikolopoulos, Nikolaos P
Using the Framework for Processing Video developed by Osama Masoud at the University of Minnesota, this study sought to identify and analyze distractions to the driver both within a vehicle as well as outside. A dashboard-mounted camera uses infrared light bursts to identify the pupil of the driver's eye. The software then tracks the relative position of the eye and pupil to make observations about the driver's gaze. The research also includes a method for measuring the driver's response to traffic and interactions between the driver and vehicle itself. The results will be used to study distractions to the driver and its affect on driver behavior in real road conditions.
Monitoring Weaving Sections
(2001-10-01) Masoud, Osama; Rogers, Scott; Papanikolopoulos, Nikolaos P
Traffic control in highway weaving sections is complicated since vehicles are crossing paths, changing lanes, or merging with through traffic as they enter or exit an expressway. There are two types of weaving sections: (a) single weaving sections which have one point of entry upstream and one point of exit downstream; and (b) multiple weaving sections which have more than one point of entry followed by more than one point of exit. Sensors which are based on lane detection fail to monitor weaving sections since they cannot track vehicles which cross lanes. The fundamental problem that needs to be addressed is the establishment of correspondence between a traffic object A in lane x and the same object A in lane y at a later time. For example, vision systems that depend on multiple detection zones simply cannot establish correspondences since they assume that the vehicles stay in the same lane. The motivation behind this work is to compensate for inefficiencies in existing systems as well as to provide more comprehensive data about the weaving section being monitored. We have used a vision sensor as our input. The rich information provided by vision sensors is essential for extracting information that may cover several lanes of traffic. The information that our system provides includes (but is not limited to): (a) Extraction of vehicles and thus a count of vehicles, (b) Velocity of each vehicle in the weaving section, and (c) Direction of each vehicle (this is actually a trajectory versus time rather than a fixed direction, since vehicles may change direction while in a highway weaving section). The end-product of this research is a portable system that can gather data from various weaving sections. Experimental results indicate the potential of the approach.
Pedestrian Control Issues at Busy Intersections and Monitoring Large Crowds
(2002-03-01) Maurin, Benjamin; Masoud, Osama; Rogers, Scott; Papanikolopoulos, Nikolaos P
The authors present a vision-based method for monitoring crowded urban scenes involving vehicles, individual pedestrians, and crowds. Based on optical flow, the proposed method detects, tracks, and monitors moving objects. Many problems confront researchers who attempt to track moving objects, especially in an outdoor environment: background detection, visual noise from weather, objects that move in different directions, and conditions that change from day to evening. Several systems of visual detection have been proposed previously. This system captures speed and direction as well as position, velocity, acceleration or deceleration, bounding box, and shape features. It measures movement of pixels within a scene and uses mathematical calculations to identify groups of points with similar movement characteristics. It is not limited by assumptions about the shape or size of objects, but identifies objects based on similarity of pixel motion. Algorithms are used to determine direction of crowd movement, crowd density, and mostly used areas. The speed of the software in calculating these variables depends on the quality of detection set in the first stage. Illustrations include video stills with measurement areas marked on day, evening, and indoor video sequences. The authors foresee that this system could be used for intersection control, collection of traffic data, and crowd control.
Recognition of Human Activity in Metro Transit Spaces
(2004-06-01) Gasser, Gillaume; Bird, Nathaniel; Papanikolopoulos, Nikolaos P
In this report, we introduce a vision-based system to monitor for suspicious human activities at a bus stop. The system currently examines behavior for drug dealing activities which is characterized by individuals loitering around the bus stop for a very long time with no intention of using the bus. To accomplish this goal, the system must measure how long individuals loiter around the bus stop. To facilitate this, the system must track individuals from the video feed, identify them, and keep a record of how long they spend at the bus stop. The system is broken into three distinct portions: background subtraction, object tracking, and human recognition. The background subtraction and object tracking modules use off-the-shelf algorithms and are shown to work well following people as they walk around a bus stop. The human recognition module segments the image of an individual into three portions corresponding to the head, torso, and legs. Using the median color of each of these regions, two people can be quickly compared to see if they are the same person.
Sensor-based Ramp Monitoring
(2003-05-01) Papanikolopoulos, Nikolaos P; Masoud, Osama; Wahlstrom, Eric
This report covers the creation of a system for monitoring vehicles in highway on-ramp queues. The initial phase of the project attempted to use a blob tracking algorithm to perform the ramp monitoring. The current system uses optical flow information to create virtual features based on trends in the optical flow. These features are clustered to form vehicle objects. These objects update themselves based on their statistics and those of other features in the image. The system has difficulties tracking vehicles when they stop at ramp queues and when they significantly occlude each other. However, the system succeeds by detecting vehicles entering and exiting ramps and can record their motion statistics as they do so. Several experimental results from ramps in the Twin Cities are presented.

University Digital Conservancy

Browse by Author

Browsing by Author "Papanikolopoulos, Nikolaos P"