A Hybrid Algorithm Of Gabor Filter and Gaussian Distribution Feature Extraction Techniques for Facial Expression Recognition
Aruna Bhadu, Hardyal Singh Shekhawat, Vijay Kumar, Tarun Kumar
Available Online April 2013.
- Facial expression recognition is necessary for designing any human-machine interfaces. Facial expression plays an important role in the recognition of human emotions and non verbal communication. Human face is a rich and powerful source of communicative information about human behavior. In this dissertation work I contribute to design a feature extraction technique. An efficient facial expression recognition method is proposed. The method uses a combine set of features obtained from Gabor Filter bank with 3 frequencies and 5 different angles and the Gaussian distribution methods. The performance of the proposed method is compared with the Gabor Filter and Gaussian distribution methods. The classification tasks perform using the Adaboost classifier. The training and testing images selected from the publicly available JAFFE (Japanese Female Facial Expression) dataset. For the experimental work 70% data are used for training purpose and remaining 30% data are used for testing purpose. The classification results show that the proposed combine feature extraction methods of Gabor filter and Gaussian distribution provide high recognition rate compared to the individual Gabor filter and Gaussian Distribution Feature Extraction method.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Aruna Bhadu AU - Hardyal Singh Shekhawat AU - Vijay Kumar AU - Tarun Kumar PY - 2013/04 DA - 2013/04 TI - A Hybrid Algorithm Of Gabor Filter and Gaussian Distribution Feature Extraction Techniques for Facial Expression Recognition PB - Atlantis Press SP - 145 EP - 149 SN - 1951-6851 UR - https://www.atlantis-press.com/article/6294 ID - Bhadu2013/04 ER -