Analysis of Front-view Face For Face Profile Recognition

• Main Author: 侯士東
• Other Authors: Ching-Tang Hsieh
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/33272334508676813246

碩士 === 淡江大學 === 電機工程學系 === 92 === Face recognition has established itself as an important subbranch of pattern recognition within the field of computer science. Many state-of-the-art systems have focused on the task of recognizing frontal views of people. In this paper we propose an approach to recognize profile face with a system trained on transformed frontal face. The features of the system consist of global feature and local features. The global feature is the eigenface of the original image. The local features are extracted from the subbands of the eyes , nose and mouth region of the original image with the wavelet transform. We use the principal component analysis (PCA) method to decrease the dimension of the local features. The Euclidean distance is used for matching. Experimental result shows that the proposed method is superior to the conventional methods.