Improve the Recognition Rate of Facial Expressions by Normalized Facial Features of Different Personal Face and Photo Sizes
Proceeding: The Fourth International Conference on Informatics & Applications (ICIA2015)Publication Date: 2015-07-20
Authors : Shih-Lin Wu; Jun-Wei Hsu;
Page : 170-174
Keywords : Facial Expression Recognition; Feature Extraction; SVM.;
Abstract
With the advancement in camera techniques, the recognition of human facial expressions has become an important research topic in recent years. Most of the face recognition systems are trying to seek facial features such as eyes and mouth automatically. But the results are inefficient because of the wrong angle and size of a face compared with facial database. In this paper, we propose a real time facial expression recognition system which possesses two functions. First, the system extracts facial features from human face and rotates to the right angle if the face did not pose the right angle. Second, to solve the different face sizes caused by near-far distance to the camera, the system employs a normalization method based on the fixed length of some features on user faces. From experimental results, the system can identify the facial expressions with a high precision rate.
Other Latest Articles
- Meaningful Object Extraction for Booking Website
- Visual Object Tracking Based on Dynamic Weights and Gaze Density
- Robust Local Graph Structure for Texture Classification
- The Application of Facial Expression Recognition to the Evaluation of Abstract Graphics
- Real-time Caption Detection and Localization in TV Programs via Stroke Width Transform and Morphological Transform
Last modified: 2015-08-10 22:21:09