College of Engineering >> Administration >> Communications and Marketing >> Seminars >> 2006 Archives >> Fall >>

Neural Networks for Computer Vision

Date: Oct. 16
Time: 2:00 p.m.
Place: 206 Tichenor Hall map

Martin Carlisle
Professor and Advisor-in-Charge
Department of Computer Science
United States Air Force Academy

Martin C. Carlisle, PhD is a professor of computer science at the United States Air Force Academy. Last year, he spent a 1 year sabbatical at Auburn University at a visiting associate professor.  His research interests include computer security, simulation, neural networks and computer science education.  He is the primary author of severalopen-source software products used worldwide, including RAPTOR, AdaGIDE and A#.  He has a Bachelor of Science in mathematics and computer science from the University of Delaware, and a Master of Arts and Doctorate in computer science from Princeton University.

Abstract

Neural Networks for Computer Vision
This talk will describe preliminary research at the US Air Force Academy to create computer programs that recognize features in images. An overview of neural networks and convolutional neural networks will be given, along with a description of both supervised and unsupervised learning strategies. Finally, results from training a convolutional neural network to recognize patterns, and then using this network to examine new images (not from the training set) will be shown, along with some ideas for future directions.