Facial emotion recognition using empirical mode decomposition

作者：

Highlights：

• A new approach of using EMD and KLFDA for facial emotion recognition is presented.

• Dimension reduction by PCA + LDA, LFDA and KLFDA can improve the performance.

• Recognition rate of 99.75% was achieved by IMF1 + KLFDA using ELM classifier.

摘要

•A new approach of using EMD and KLFDA for facial emotion recognition is presented.•Dimension reduction by PCA + LDA, LFDA and KLFDA can improve the performance.•Recognition rate of 99.75% was achieved by IMF1 + KLFDA using ELM classifier.

论文关键词：Facial emotion,Empirical mode decomposition,PCA + LDA,LFDA,KLFDA,Extreme learning machines

论文评审过程：Available online 16 September 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.08.049