Publication detail

Combining Features for Recognizing Emotional Facial Expressions in Static Images

PŘINOSIL, J. SMÉKAL, Z. ESPOSITO, A.

Original Title

Combining Features for Recognizing Emotional Facial Expressions in Static Images

English Title

Combining Features for Recognizing Emotional Facial Expressions in Static Images

Type

journal article in Web of Science

Language

en

Original Abstract

This work approaches the problem of recognizing emotional facial expressions in static images focusing on three preprocessing techniques for feature extraction, such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Gabor filters. These methods are commonly used for face recognition and the novelty consists in combining features provided by them in order to improve the performance of an automatic procedure for recognizing emotional facial expressions. Classification performance experiments, testing new expressions and new subjects, were performed on the Japanese Female Facial Expression (JAFFE) database using a Multi-Layer Perceptron (MLP) Neural Network as classifier. The best classification performance on new expressions was obtained combining PCA and LDA features (93% of correct recognition rate), whereas that on new subjects was obtained combining PCA, LDA and Gabor filter features (94% of correct recognition rate).

English abstract

This work approaches the problem of recognizing emotional facial expressions in static images focusing on three preprocessing techniques for feature extraction, such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Gabor filters. These methods are commonly used for face recognition and the novelty consists in combining features provided by them in order to improve the performance of an automatic procedure for recognizing emotional facial expressions. Classification performance experiments, testing new expressions and new subjects, were performed on the Japanese Female Facial Expression (JAFFE) database using a Multi-Layer Perceptron (MLP) Neural Network as classifier. The best classification performance on new expressions was obtained combining PCA and LDA features (93% of correct recognition rate), whereas that on new subjects was obtained combining PCA, LDA and Gabor filter features (94% of correct recognition rate).

Keywords

Principal Component Analysis, Linear Discriminant Analysis, Gabor filters, facial features, basic emotions.

RIV year

2008

Released

12.12.2008

Publisher

Springer

Location

Berlin

Pages from

59

Pages to

72

Pages count

13

BibTex


@article{BUT49151,
  author="Jiří {Přinosil} and Zdeněk {Smékal} and Anna {Esposito}",
  title="Combining Features for Recognizing Emotional Facial Expressions in Static Images",
  annote="This work approaches the problem of recognizing emotional facial expressions in static images focusing on three preprocessing techniques for feature extraction, such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Gabor filters. These methods are commonly used for face recognition and the novelty consists in combining features provided by them in order to improve the performance of an automatic procedure for recognizing emotional facial expressions. Classification performance experiments, testing new expressions and new subjects, were performed on the Japanese Female Facial Expression (JAFFE) database using a Multi-Layer Perceptron (MLP) Neural Network as classifier. The best classification performance on new expressions was obtained combining PCA and LDA features (93% of correct recognition rate), whereas that on new subjects was obtained combining PCA, LDA and Gabor filter features (94% of correct recognition rate).",
  address="Springer",
  chapter="49151",
  institution="Springer",
  journal="Lecture Notes in Computer Science (IF 0,513)",
  number="5042",
  volume="2008",
  year="2008",
  month="december",
  pages="59--72",
  publisher="Springer",
  type="journal article in Web of Science"
}