Facial expression classification using Zernike Moment Invariant and Artificial Neural Network

摘要：

In this paper, we introduce a novel method to discriminate seven prototypic facial expressions. The new method employs the Feature Zernike Moment Invariant and Artificial Neural Network (FZMI-ANN), which is a human face expression classification system. In our research, face image was projected in an appropriate Zernike Moment invariant transform method; a constructive procedure was detailed and a systematic performance on a public database Japanese Female Facial Expression (JAFFE) was evaluated. From our experimental results that have demonstrated the potential capabilities of the proposed method, further use of ANN for facial expression classification based on local features extracted by FZMI technique has been recommended.

关键词： Zernike moment invariant artificial neural network facial expression classification

作者: Tran Binh Long Le Hoang Thai Tran Hanh

作者单位: Department of Computer Science University of LacHong 10 HuynhVanNghe, DongNai 71000, VietNam Department of Computer Science HoChiMinh City University of Science 227NguyenVanCu, HoChiMinh 70000,

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 221-224

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）