Analysis of PCA and LDA Features for Facial Expression Recognition Using SVM and HMM Classifiers

  • Satishkumar Varma
  • Megha Shinde
  • Satishkumar S. Chavan
Conference paper


The face of a person is full of emotions. The emotions can be used to interact between human beings and computers. Different systems have been developed to recognize emotions. In this paper, a facial expression recognition system has been developed to recognize emotions which have four steps mainly face image acquisition, pre-processing, feature extraction and face recognition with six emotions. In the first step, a single face image was selected as an input. In the second step, this face image was converted into HSV color space and median filtering was used to remove the noise. Two methods namely Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) have been used for feature extraction and representation in the third step. During the fourth step, the Support Vector Machine (SVM) and Hidden Markov Model (HMM) have been used to classify the given face over two different datasets. Finally, the experiments were performed by combining two feature vectors to recognize the face and the proposed system is able to recognize six basic emotions viz. happiness, sadness, anger, disgust, surprise and fear. The performance has been evaluated and presented using confusion matrix and classification accuracy. The combination of features from PCA and LDA with SVM is effective and useful for classification and recognition of facial expressions.


Facial expression recognition Emotion classification Principal component analysis Linear discriminant analysis Support vector machine Hidden Markov model 


  1. 1.
    Bakshi U, Singhal R (2014) A survey on face detection methods and feature extraction techniques of face recognition. Int J Emerg Trends Technol Comput Sci 3(3):233–237Google Scholar
  2. 2.
    Kaur A, Singh S (2015) Taqdir: face recognition using PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis) Techniques. Int J Adv Res Comput Commun Eng 4(3):308–310CrossRefGoogle Scholar
  3. 3.
    Martinez A, Kak A (2001) PCA versus LDA. IEEE Trans Pattern Anal Mach Intell 23(2):228–233CrossRefGoogle Scholar
  4. 4.
    Dhall S, Sethi P (2014) Geometric and appearance feature analysis for facial expression recognition. Int J Adv Eng Technol 5(3):1–11Google Scholar
  5. 5.
    Turk M, Pentland A (1991) Eigenfaces for recognition. J Cogn Neurosci 3(1):71–86CrossRefGoogle Scholar
  6. 6.
    Hameed Siddiqi M, Ali R, Kha AM (2015) Human facial expression recognition using linear discriminant analysis and hidden Markov model. IEEE Trans Image Process 24(4):1386–1398MathSciNetCrossRefGoogle Scholar
  7. 7.
    Shan C, Gong S, McOwan PW (2009) Facial expression recognition based on Local Binary Patterns: a comprehensive study. J Image Vision Comput 27(6):803–817CrossRefGoogle Scholar
  8. 8.
    Gosavi AP, Khot SR (2013) Facial expression recognition using principal component analysis. Int J Soft Comput Eng 3(4):258–262Google Scholar
  9. 9.
    Bouzalmat A, Kharroubi J, Zarghili A (2013) Face recognition using SVM based on LDA. Int J Comput Sci 10(4):171–179Google Scholar
  10. 10.
    Nefian AV, Hayes MH (1998) Hidden Markov models for face recognition. In: International conference on image processing, vol 5, pp 2721–2724, IEEEGoogle Scholar
  11. 11.
    Delac K, Grgic M, Grgic S (2005) Independent comparative study of PCA, ICA, and LDA on the FERET data set. Int J Imaging Syst Technol 15(5):252–260CrossRefGoogle Scholar
  12. 12.
    Weston J, Watkins C (1999) Support vector machines for multi-class pattern recognition. In: Verleysen M (ed) Seventh European symposium on artificial neural networks (ESANN), pp 219–224Google Scholar
  13. 13.
    Chapelle O, Vapnik V (1999) Model selection for support vector machines. In: Proceeding advances in neural information processing systemsGoogle Scholar
  14. 14.
    Lee JH, Lin CJ 2000) Automatic model selection for SVM. Technical report, Department of Computer Science & Information Engineering, National Taiwan UniversityGoogle Scholar
  15. 15.
    Hsu CW, Chang CC, Lin CJ (2003) A practical guide to support vector machines. Technical report, Department of Computer Science & Information Engineering, National Taiwan UniversityGoogle Scholar
  16. 16.
    Staelin C (2002) Parameter selection for support vector machines. Technical report, HP Labs, Israel, pp 1–4Google Scholar
  17. 17.
    Libor Spacek’s Facial Images Databases, Computer Vision Science Research Projects (2009). URL:

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Satishkumar Varma
    • 1
  • Megha Shinde
    • 1
  • Satishkumar S. Chavan
    • 2
  1. 1.Pillai College of EngineeringNavi MumbaiIndia
  2. 2.Don Bosco Institute of TechnologyMumbaiIndia

Personalised recommendations