Welcome to Kwon's Research Page
This page is intended to provide a brief overview of my current research work at UMD. Unfortunately, it seems, I could never find enough time to construct a good web page that I wish to provide. It will be always partial, but if anyone is interested in my research beyond what is available here, please don't hesitate
to contact me. 218-726-8211. Link to Dr. Kwon's vita .
My Recent Funded Projects:
US DOT IVI (Intelligent Vehicles Initiative) Specialty Vehicle Field Operational Test
The project was developed based on a collaborated team effort comprising Mn/DOT, the University of Minnesota, 3M, who provides a magnetic based lateral guidance system, Navistar, a supplier of medium and heavy duty trucks, and Altra Technologies, an Eden Prairie, MN, based provider of radar based collision warning system. The project goal is to develop an optimized user interface integrating both vehicle guidance and collision avoidance technologies and to document the safety and performance benefits associated with such systems by performing a field operational test. My role in this project is to evaluate motorist’s visibility through weather sensor network and video images collected during the test period. This project was funded by multiple agencies including US DOT, Mn/DOT and private sectors with total funding of $6.5 million.
Northland Advanced Transportation Systems Research Lab (NATSRL)
The geographical location of the Duluth and Northern Minnesota area presents unique challenges in operating and maintaining the area-wide transportation systems, which need to provide safe and efficient travel environment under often inclement weather conditions. NATSRL was embarked on to address such research needs for Northland transportation systems. The research issues include efficient real-time detection of hazardous road/weather/traffic conditions, advance warning and guidance to drivers with road/weather/traffic information, winter-road and snowplow fleet management and decision support system with on-line weather/pavement information, large-scale event and tourist traffic management, and on/off-line assessment of area-wide system performance. This project was funded by US DOT TEA-21, Mn/DOT, and UMD, about $4.0 million. .
TMC Traffic Data Automation For Mn/DOT’s Traffic Monitoring Program
The Minnesota Department of Transportation (Mn/DOT) has been responsible for collecting, analyzing, and publishing traffic count data from the various roadway systems throughout the state. The traffic reporting system mainly developed by the Traffic Forecasting and Analysis Section (TFAS) of Mn/DOT has been used in several federal programs, internal Mn/DOT applications, and many private sectors. The objective of this project is to develop computerized automation methods for the current manual effort to import, filter, and analyze the Mn/DOT TMC portion of inductive loop detector data contributed to the Mn/DOT's Traffic Monitoring System. The main research effort has been in developing multi-level data screening and analysis methods based prioritized acceptance tests and Q-K curve analysis. This project was sponsored by the Mn/DOT Guide Star Program. $79,819.
Visibility Measurement System Based on Imaging Technologies
This research project focuses on developing a practical atmospheric-visibility monitoring system based on imaging systems. Because visibility reductions due to inclement weather conditions are one of the main causes of traffic incidents, and among one of the primary criteria used to determine road closures in winter, accurate visibility measurement is prime importance to many transportation decision makers. However, accurate and reliable measurement of atmospheric visibility is challenging because it continuously changes over time and space and is influenced by a host of atmospheric conditions such as fog, rain, snow, smog, dust, sun direction, solar radiation, etc. Measurement by human observers is often unreliable due to differences in individual eyesight, perception and other biological conditions. Other techniques such as light-scatter meters exist, but do not measure the true visibility. A new approach developed in this research is based on the measurement of visibility through an imaging system that comprises a surveillance video camera, an image digitizer and multiple targets positioned at specific distances from the camera. For daytime, an image-processing algorithm was developed to determine at what distance the foreground is no longer distinguishable, by which the distance represents the visibility. For night visibility, several approaches are under investigation. Those include embedding light sources in the targets with the frequency range in visual and infrared spectrum and using different types of spectral filters in the camera. We also investigate a new way of measuring visibility based on the concept of relativity. This research was sponsored by US DOT TEA-21 and Mn/DOT. $93,191.
Click here to see a daytime visibility screen of the system.
Click here to see a night visibility screen of the system.
Final report Phase-1: Atmospheric Visibility Measurements Using Video Cameras
Final report Phase-2: Atmospheric Visibility Measurements Using Video Cameras: Relative Visibility
Next-Generation Road Weather Information System: Concept and Prototype Development
Traditional ways of using R/WIS (Road/Weather Information System) have been to forecast road icing before its formation for proactive winter-road maintenance. Although ice forecasting has served a significant value in deicing operations, as states plan implementation of a statewide R/WIS, they recognize that it is a large, expensive system. Moreover, due to rapidly changing technologies and associated increasing costs, implementation of R/WIS has been complex. Planners find that it is important to start with an architecture that allows easy integration of new and legacy technologies, and then gradual expanding to a larger system. This project develops a flexible R/WIS model that can seamlessly integrate with the existing and legacy systems and can gradually expand to a larger system. The research also focuses on new ways of utilizing the massive amount of environmental sensor data collected from RWIS. This research was sponsored by Mn/DOT.
Web-Based Pavement Condition Reporting System
Pavement Condition Reporting System (PCRS) is a Mn/DOT’s networked application tool that tracks and reports weather circumstances that affect driving conditions. The original PCRS was designed based on a database server and customized front-end programs that allow data entry and retrieval. The main problem of this system was the customized front-end programs that required recoding as the client operating systems were upgraded. Indeed, the system did not work as Mn/DOT upgraded their operating systems. However, even if the old software was rewritten for new operating systems with extra budget, modification of the front- end programs required reinstallation of the codes in every client workstations in order for the Mn/DOT PCRS to work properly. In order to alleviate the problems associated with the old PCRS and to further advance the PCRS technology, Mn/DOT decided to implement a new PCRS based on a World Wide Web (web) technology. In this new approach, any Mn/DOT operators having a standard web browser and a password would have access to the PCRS for all of its functions without the needs for software update. This project was given to Dr. Kwon and his students at the University of Minnesota Duluth, and they successfully developed a new one, referred to as web-PCRS, using the state-of-the-art dynamic web technologies. The system was driven by the sever-side codes of the central web-server integrated by a centralized relational database. It includes various forms of pavement-condition data entry, real-time notification, instant messaging between PCRS operators and administrators, weather/permit entry, and many administrative functions that would be controllable from the client side. The developed system was deployed statewide. This project was sponsored by Mn/DOT, $44,000.
Development of a Mobile Thermal Mapping System for Mn/DOT
A system that gathers information from road weather sensor stations
and utilizes it in forecasting future road conditions for road maintenance,
is referred to as the Road Weather Information System (RWIS).
Early experiments and experiences in European countries suggest that
the RWIS helps road maintenance crews for efficient snow control
and ice prevention. The Minnesota Department of Transportation (Mn/DOT)
RWIS committee carefully reviewed literature,
economics and the European RWIS experiences, and recommended the establishment
of a state wide RWIS network in the Minnesota state (RWIS committee, 1994).
The aim of the RWIS
network is twofold. First,
the RWIS network is designed to predict future road surface conditions based
on the information of the real-time data collected from the RWIS sensor
stations, and
the weather forecast obtained from national or local weather services.
This forecast can then be directly forwarded to road maintenance crews
for snow control and ice prevention, and to vehicle drivers for road driving
information.
Secondly, it works as a
road-condition monitoring station at frequent problem areas, by which
emergent road conditions are early detected, and thus
a prompt maintenance service is called for.
The overall RWIS state-wide system consists of two major components:
The stationary
RWIS sensor stations and the thermal-map surveys of road ways.
The RWIS sensor stations report their recorded road conditions
to the central computer (about every five or
ten minutes), where a prediction model computes and forecasts
future road conditions based on the past thermal-map surveys and
the real time data received from the RWIS sensor
stations.
Therefore, thermal-map survey is an extremely important step as an
integrated part of the RWIS network.
At initial stages of the RWIS network construction, thermal-map survey is
also used to locate the optimal RWIS sites and to determine
the number of required stations.
In this project, I designed and implemented a mobile thermal mapping
instrument using a Ford Astro ban. This system has been used to collect
data from 1995-96 winter period and has been scheduled for the next
winter. This project was funded by Mn/DOT MORE fund.
The following areas of research are not funded, but under way due to my own
interests.
Global Optimization Neural Network
One-Pass Learning Neural Network
Non-Uniform Image Processing
Most of the techniques employed in the area of image processing have been
explored by uniform processing approaches in its nature.
For example, image compression uniformly compresses all areas of the image
without regarding such as the human perception of objects in the image (which
may provide critical information on "which portion of the image is more
interested or important than others").
Another such area is in image interpolation.
Most interpolation techniques apply one formulation or another without knowing
how much distortion would be caused to the image as a result of this uniform
formulation of the interpolation algorithm, which essentially leads to
introducing many blocky effects.
The effect on the Fourier Transform is a non-issue and at times it is very
cumbersome to find the effects of these arbitrary interpolation techniques
on the frequency domain. This creats havoc on the frequency components.
A better and sensible approach is associating a non-uniform processing
based on human perception. My research effort is to develop a non-uniform image processing that accounts human perception in particular for image compression
and interpolation.
Object Recognition
The area of object recognition has traditionally been synonymous with the
technique of template matching. However I believe that it is highly unreliable
due to some of its inherent characteristics. My effort is to
develop a newer technique that can be more generalized in terms of human intelligence and perception.
The direction of this research is to follow the simple common-sense
concept that we human beings detect or recognize a pattern or an object from
a scene by interpreting the syntax of features that make up an object.
This approach is incrediably efficient in terms of data as demonstrated by few
strokes of cartoon artists.
Unfortunately, we know very little about
how to automatically extract features and feature-syntax relation from raw data.
I am presently in the process of developing few new techiques and applying
the algorithm in visibility measurements for the Deaprtment of Transportation.
Back to my home page
Kwon's home Page.
Back to the page2 page
