Capturing Eye State (Open/Closed) by Image Analysis for Drowsiness
 Detection
 
 Dion Monstavicious, Benjamin Blizard, Nathan Eagle.
 
 Feature recognition in human faces has many important applications.  In
 this project, we will investigate methods to detect eyes in pictures
 containing human faces.  Our first priority is to reliably detect if a
 person's eyes are open or closed from a picture of the face.  Although
 many applications are possible, this algorithm would be a central
 component of a computer vision system used to detect sleepy operators of
 motor vehicles.  Time permitting, we will investigate other possible
 applications of an eye detector (ie: head position and orientation) and
 attempt to derive blink rate from a video stream.
 
 Our Primary Objectives:
 1) search an image for a human face using previously developed algorithms,
 such as the Rowley-Baluja-Kanade Face Detector (or perhaps a previous
 project implementation if available).
 
 2) search the face region for eyes and detect if the eyes are open or
 closed. This may be some sort of pattern recognition, feature extraction,
 or a neural network algorithm (modification of face detector for eyes).
 
 
 References:
 
 M. Yang, D. Kriegman, N. Ahuja, Detecting Faces in Images: A Survey,
 Department of Computer Science and Beckman Institute Technical Monograph,
 University of Illinois at Urbana-Champaign, Urbana IL, 61801
 
 H. Rowley, S. Baluja, and T. Kanade, "Neural Network-Based Face
 Detection," IEEE Transactions on Pattern Analysis and Machine
 Intelligence, Vol. 20, No. 1, January, 1998, pp. 23-38.
 
 Sanner, Scott, "CS223B Winter quarter, Final Project"
 http://www.stanford.edu/~sanner/Vision/Project.html