Emotion recognition using Principal Component Analysis with Singular Value Decomposition

Author

Gosavi, Ajit P. ; Khot, S.R.

Author_Institution

Electron. & Telecommun., D.Y. Patil Coll. of Eng. & Technol., Kolhapur, India

fYear

2014

fDate

13-14 Feb. 2014

Firstpage

1

Lastpage

5

Abstract

Emotion recognition plays vital role in Human Computer Interface. This paper focuses on facial expression to identify seven universal human emotions such as, happy, disgust, neutral, anger, sad, surprise and fear. This is carried out by trying to extract unique facial expression features among emotions using Principal Component Analysis with Singular Value Decomposition and Euclidean Distance Classifier. Using public database Japanese Female Facial Expression (JAFFE) recognition is obtained nearly 78.57%. Recognition rate and Accuracy of various expressions using Principal Component Analysis alone and Principal Component Analysis with Singular Value Decomposition is compared.

Keywords

emotion recognition; face recognition; feature extraction; image classification; principal component analysis; singular value decomposition; Euclidean distance classifier; JAFFE; Japanese Female Facial Expression; emotion recognition; human computer interface; principal component analysis; singular value decomposition; unique facial expression features; universal human emotions; Accuracy; Databases; Emotion recognition; Face recognition; Principal component analysis; Target recognition; Facial Expression Detection; Feature Extraction; Principal Component Analysis (PCA); Singular Value Decomposition (SVD);

fLanguage

English

Publisher

ieee

Conference_Titel

Electronics and Communication Systems (ICECS), 2014 International Conference on

Conference_Location

Coimbatore

Print_ISBN

978-1-4799-2321-2

Type

conf

DOI

10.1109/ECS.2014.6892683

Filename

6892683