Mobile Embedded System Implementation for Sleep Detection

• Main Authors: Tsung-Kuang Lee, 李宗光
• Other Authors: 謝君偉
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/18668584900920971858

碩士 === 元智大學 === 電機工程學系 === 97 === Our research is focus on sleeping detection of driver on embedded system. We use Ada-Boosting algorithm based on the feature of Haar Basics to extract the face feature and locate the possible face regions. The detection speed can be improved by using a cascade classifier method. For sleeping detection, our system applies the morphology operation to enhance feature and remove the noise. Then, we use a moment threshold to obtain the binary information of a face region. A connected component is used to locate the possible eye positions. According to the symmetry and position, the sleeping situation can be checked.