Our initial project idea was to develop a computer algorithm that would be able to locate Waldo from an image in a Where's Waldo Search books. However, after further investigation we realized that this project had been previously attempted. During this whole ordeal our emotions had been on a roller coaster ride desperately searching for a project. Shortly after we thought of how convenient it would have been if the computer we had been searching on could have read our emotions and provided us with some aid. This wishful thinking is what inspired the idea of emotion detection for the theme of our project. After consulting, with Chris Rozell and Courtney Lane we decided the best approach was to develop an algorithm that took in a few images of a person and classified the images according to the emotions expressed in the image. After developing this plan of attack we brainstormed many different ways to detect emotions from images. Much of online resources had took a computer science approach to solving the same task at hand. After a bit of anxiety we took a dive in and developed programs that could process an image just as a human would go about. We began by looking for edges and also computed different wavelets hoping to find a vast distinction between emotions. The four images we decided to investigate where happy, surprised, angry, and sad. After several different approaches we found the best way to solve our problem was to use a top down method discussed in the approach section. For further more technical explanation continue reading with the problem.

Second half of intensive survey of brain and behavioral studies for first-year graduate students. Topics include: perception, attention, working memory, recognition and recall, categorization, language, and thought. Topics covered from behavioral, computational, and neural perspectives.

This class is the second half of an intensive survey of cognitive science for first-year graduate students. Topics include visual perception, language, memory, cognitive architecture, learning, reasoning, decision-making, and cognitive development. Topics covered are from behavioral, computational, and neural perspectives.

This course will serve as both an introduction to contemporary political philosophy and a way to explore issues of pluralism and multiculturalism. Racial and ethnic groups, national minorities, aboriginals, women, sexual minorities, and other groups have organized to highlight injustice and demand recognition and accommodation on the basis of their differences. In practice, democratic states have granted a variety of group-differentiated rights, such as exemptions from generally applicable laws, special representation rights, language rights, or limited self-government rights, to different types of groups. This course will examine how different theories of citizenship address the challenges raised by different forms of pluralism. We will focus in particular on the following questions: - Does justice require granting group-differentiated rights? - Do group-differentiated rights conflict with liberal and democratic commitments to equality and justice for all citizens? - What, if anything, can hold a multi-religious, multicultural society together? Why should the citizens of such a society want to hold together?

Seminar covering topics of current interest in biology. Includes reading and analysis of research papers and student presentations. Contact Biology Education Office for topics.

Training and testing of our instrument recognition system.

Once the Eigenfaces are computed, one can use the vectors to detect new faces images into the system.

An introduction to automated face recognition using eigenfaces.

A description of the database used in digit identification.

Experimental setup for acquiring images.

Pacing guides for The Overland Trail unit of IMP Year 1 (2nd edition)

Process of converting an image of handwritten digits into a standardized format.

The Eigenface method for face recognition is currently the fastest and most effective method known. This project implements and tests the method on several image sets in order to determine the characteristics of the eigenface method.

This module provides an overview of Team SSB's final ELEC 301 project -- accent recognition and classification with a neural network

A project by Rice University students done in Fall 2004 for ELEC 301. An investigation of signal processing techniques applied to iris recognition.

Results of the iris recognition program and conclusions about the technology.

Listed in this module are output results and analysis of test-audio samples from various speakers, fed to a trained neural network.

How to scale pitch and instrument recognition in mophonic recordings to polyphonic recordings.

This module sums up the results of our audio file recognition system.