OSU Navigation Bar

The Ohio State University University Libraries Knowledge Bank

A User Study Contrasting 2D Unsteady Vector Field Visualization Techniques

Please use this identifier to cite or link to this item: http://hdl.handle.net/1811/300

Show full item record

Files Size Format View
NathanAndrysco.pdf 1.627Mb PDF View/Open

Title: A User Study Contrasting 2D Unsteady Vector Field Visualization Techniques
Creators: Andrysco, Nathan
Advisor: Shen, Han-Wei
Issue Date: 2005-06
Abstract: Scientific visualization is very popular today among many different fields. Its purpose is to take data and represent it visually so that a person can better understand what is occurring with the data. This idea can be as simple as a pie chart or as complex as drawing the beating human heart with data taken from a MRI scan. One particular area of scientific visualization is the representation of flow data. Many methods have been researched and published on how to visualize different types of flow fields, but only a few papers have been published that have attempted to compare and contrast the effectiveness of the various methods. For the study, five different flow field visualization methods are compared – LIC, pathline, streamline, streakline, and hedgehog. A framework was created to not only display each of the methods but also allow the user to interact with them. The user is shown an unsteady flow field using one of the five visualization methods. He may view the flow field at any time step using a slider or hitting a button to animate the flow. The user is asked to make a response on two different types of tests. The first one asks the user to advect a particle to a circle. The angular error is recorded into a file. The second test asks the user to determine where a massless particle might have begun. The number of clicks it takes the user to correctly identify where the particle began is recorded. Statistical analysis was performed on the data to determine which visualization method is the “best”. Unfortunately after using 95% confidence intervals, no statistical conclusions could be drawn about what were the better methods. However, it could be stated that the test method could be refined to produce better results. Or if the test method is fine, it just might be the fact that even with the best visualization methods out there that a person is unable to fully comprehend an unsteady flow field.
Series/Report no.: The Ohio State University. Department of Computer Science and Engineering Honors Theses;2005
Keywords: Flow Visualization
User Study
Unsteady Flow Visualization
Vector Visualization
URI: http://hdl.handle.net/1811/300
Bookmark and Share