University of Minnesota
University Relations
http://www.umn.edu/urelate
612-624-6868
myU OneStop


Go to unit's home.

Home | DTI | 2005–06 funded proposals | Ellen K. Longmire, Ivan Marusic, Nikos Papanikolopoulos

Initiatives in Digital Technology: 2005–06 Funded Proposals

Ellen K. Longmire, Ivan Marusic, Nikos Papanikolopoulos

Computer Vision Methods for Understanding Turbulent Flows: The proposed cross-disciplinary research unites experts in experimental fluid mechanics and computer vision to tackle the problem of understanding the eddy structure of turbulent flows.

The proposed cross-disciplinary research unites experts in experimental fluid mechanics and computer vision to tackle the problem of understanding the eddy structure of turbulent flows. These flows, which are of great importance in a large number of practical applications, have remained a challenging area of research for over 100 years. The project will focus on development of computer vision methods appropriate for identifying important features within turbulent wall-bounded flows. Specifically, we seek to identify signatures of superstructures that dominate the energetics and transport within the flow such as groups of vortices that travel as packets. Methods to be examined include dynamic contour, skeletonization, and curve evolution algorithms. The methods will be tested against experimental data from dual-plane PIV (particle image velocimetry) fields acquired in a turbulent boundary layer and numerical data from a direct numerical simulation of turbulent channel flow. Vortex packets as well as other potential superstructures are characterized by a number of parameters, and the challenge will be to utilize combinations of these parameters to advantage within computer vision algorithms.

The research proposed will explore and demonstrate novel analysis tools for the study of turbulence. These tools can be applied to probe both experimental and numerically-simulated data to gain new insight into the dominant structures in wall-bounded turbulence. Such insight has the potential to lead to new drag reduction strategies for vehicles as well as more accurate predictive codes for practical applications in the environment and the aerospace, transportation, energy, and chemical processing industries. Additional broad impacts that will result from the proposed work include interaction of the funded graduate students with other graduate and undergraduate students in the relevant laboratories. All three PIs have worked with many undergraduate researchers through NSF REU, the CRA-W distributed mentor project, and UROP.