LOCAL FEATURE SELECTION USING THE WRAPPER APPROACH FOR FACIAL-EXPRESSION RECOGNITION

(Received: 15-Apr.-2024, Revised: 29-Jun.-2024 , Accepted: 16-Jul.-2024)

Authors Fatima Zohra Boukhobza, Abdenour Hacine Gharbi, Khaled Rouabah, Phillipe Ravier,

Keywords #Facial expression recognition #Local feature extraction #Feature selection #Wrapper approach #Classification

Abstract Automatic Facial Expression Recognition (FER) systems provide an important way to express and interpret the emotional and mental states of human beings. These FER systems transform the facial image into a set of features to train a classifier capable of distinguishing between different classes of emotions. However, the problem often posed is that the extracted feature vectors sometimes contain irrelevant or redundant features, which decreases the accuracy of the induced classifier and increases the computation time. To overcome this problem, dimensionality must be reduced by selecting only the most relevant features. In this paper, we study the impact of adding the "Wrapper" selection approach and using the information provided by different local regions of the face such as the mouth, eyes and eyebrows, on the performance of a traditional FER system based on a local geometric feature-extraction method. The objective here is to test and analyze how this combination can improve the overall performance of the original traditional system. The obtained results, based on the Multimedia Understanding Group (MUG) database, showed that the FER system combined with the proposed feature-selection strategy gives better classification results than the original system for all four classification models; namely, K-Nearest Neighbor (KNN) classifier, Tree classifier, NB classifier and Linear Discriminant Analysis (LDA). Indeed, a considerable reduction (up to 50%) in the number of features used and an accuracy of 100%, using the LDA classifier, were observed, which represents a significant improvement in terms of computation time, efficiency and memory space. Furthermore, the majority of relevant features used are part of the "eyebrows’ region", which proves the importance of using information from local regions of the face in emotion recognition tasks.

References

[1] S. Li and W. Deng, "Deep Facial Expression Recognition: A Survey," IEEE Transactions on AffectiveComputing, vol. 13, no 3, pp. 1195-1215, 2020.

[2] J. F. Cohn and F. De la Torre, "Automated Face Analysis for Affective Computing," The OxfordHandbook of Affective Computing, pp. 131-150, 2015.

[3] M. B. Abdulrazaq, M. R. Mahmood, S. R. Zeebaree, M. H. Abdulwahab, R. R. Zebari and A. B. Sallow,"An Analytical Appraisal for Supervised Classifiers’ Performance on Facial Expression Recognition Based on Relief-F Feature Selection," Journal of Physics: Conference Series, IOP Publishing, vol. 1804, no. 1, p. 012055, February 2021.

[4] M. R. Mahmood, "Two Feature Selection Methods Comparison Chi-Square and Relief-F for FacialExpression Recognition," Journal of Physics: Conference Series, IOP Publishing, vol. 1804, no. 1, pp. 012056, February 2021.

[5] A.A. Ewees, H. A. ElLaban and R. M. ElEraky, "Features Selection for Facial Expression Recognition,"Proc. of the 2019 10th IEEE Int. Conf. on Computing, Communication and Networking Technologies (ICCCNT), pp. 1-6, Kanpur, India, July 2019.

[6] M. J. Cossetin, J. C. Nievola and A. L Koerich, "Facial Expression Recognition Using a Pairwise FeatureSelection and Classification Approach," Proc. of the 2016 IEEE Int. Joint Conf. on Neural Networks (IJCNN), pp. 5149-5155, Vancouver, BC, Canada, July 2016.

[7] H. I. Dino and M. B. Abdulrazzaq, "A Comparison of Four Classification Algorithms for FacialExpression Recognition," Polytechnic Journal, vol. 10, no. 1, pp. 74-80, 2020.

[8] Y. Wang, Y. Lu and X. Wan, "Multi-feature Fusion Based on RV Correlation Coefficient for FacialExpression Recognition," Journal of Physics: Conference Series, IOP Publishing, vol. 2562, no. 1, p. 012027, August 2023.

[9] B. Richhariya and D. Gupta, "Facial Expression Recognition Using Iterative Universum Twin SupportVector Machine," Applied Soft Computing, vol.76, pp. 53-67, 2019.

[10] F. M. Talaat, Z. H. Ali, R. R. Mostafa and N. El-Rashidy, "Real-time Facial Emotion Recognition ModelBased on Kernel Autoencoder and Convolutional Neural Network for Autism Children," Soft Computing, vol. 28, pp. 6695-6708, 2024.

[11] C. Silva, A. Sobral and R. T. Vieira, "An Automatic Facial Expression Recognition System Evaluatedby Different Classifiers," Proc. of the X Workshop de Visão Computacional, Uberlândia, Minas Gerais, Brazil, pp. 208-212, October 2014.

[12] T. F. Cootes, G. J. Edwards and C. J. Taylor, "Comparing Active Shape Models with Active AppearanceModels," Proc. of the British Machine Vision Conference (BMVC), vol. 99, no. 1, pp. 173-182, September 1999.

[13] J. Li, K. Cheng, S. Wang, F. Morstatter, R. P. Trevino, J. Tang and H. Liu, "Feature Selection: A DataPerspective," ACM Computing Surveys (CSUR), vol. 50, no. 6, pp. 1-45, 2017.

[14] G. Chandrashekar and F. Sahin, "A Survey on Feature Selection Methods," Computers & ElectricalEngineering, vol. 40, no. 1, pp. 16-28, 2014.

[15] J. R. Quinlan, "Induction of Decision Trees," Machine Learning, vol. 1, pp. 81-106, 1986.

[16] Y. Saeys, I. Inza and P. Larranaga, "A Review of Feature Selection Techniques in Bioinformatics,"Bioinformatics, vol. 23, no. 19, pp. 2507-2517, 2007.

[17] P. Viola and M. J. Jones, "Robust Real-time Face Detection," International Journal of Computer Vision,vol. 57, no. 2, pp. 137-154, 2004.

[18] X. Zhao and S. Zhang, "A Review on Facial Expression Recognition: Feature Extraction andClassification," IETE Technical Review, vol. 33, no. 5, pp. 505-517, 2016.

[19] P. Ekman, "Facial Expressions," Handbook of Cognition and Emotion, vol. 16, no. 301, e320, 1999.

[20] J. N. Bassili, "Facial Motion in the Perception of Faces and of Emotional Expression," Journal ofExperimental Psychology: Human Perception and Performance, vol. 4, no. 3, pp. 373, 1978.

[21] H. Xiong, D. Zhang, C. J. Martyniuk, V. L. Trudeau and X. Xia, "Using Generalized Procrustes Analysis(GPA) for Normalization of cDNA Microarray Data," BMC Bioinformatics, vol. 9, no. 1, pp. 1-13, 2008.

[22] T. Cover and P. Hart, "Nearest Neighbor Pattern Classification," IEEE Transactions on InformationTheory, vol. 13, no. 1, pp. 21-27, 1967.

[23] D. D. Lewis, "Naive (Bayes) at Forty: The Independence Assumption in Information Retrieval," Proc. of the European Conf. on Machine Learning, pp. 4-15, Berlin, Heidelberg: Springer Berlin Heidelberg, April 1998.

[24] T. Hastie, R. Tibshirani, J. H. Friedman and J. H. Friedman, The Elements of Statistical Learning: DataMining, Inference and Prediction, vol. 2, pp. 1-758, eBook ISBN: 978-0-387-84858-7, New York: Springer, 2009.

[25] N. Suguna and K. Thanushkodi, "An Improved K-Nearest Neighbor Classification Using GeneticAlgorithm," International Journal of Computer Science Issues, vol. 7, no. 2, pp. 18-21, 2010.

[26] P. Xanthopoulos, P. M. Pardalos and T. B. Trafalis, "Linear Discriminant Analysis," Robust DataMining, Springer, pp. 27-33, Part of the Book Series: Springer Briefs in Optimization, New York, NY, 2013.

[27] B. Charbuty and A. Abdulazeez, "Classification Based on Decision Tree Algorithm for MachineLearning," Journal of Applied Science and Technology Trends, vol. 2, no. 01, pp. 20-28, 2021.

[28] S. Khalid, T. Khalil and S. Nasreen, "A Survey of Feature Selection and Feature Extraction Techniquesin Machine Learning," Proc. of the 2014 IEEE Science and Information Conf., pp. 372-378, London, UK, August 2014.

[29] J. Miao and L. Niu, "A Survey on Feature Selection," Procedia Computer Science, vol. 91, pp. 919-926.2016.

[30] S. Liu and M. Motani, "Improving Mutual Information Based Feature Selection by Boosting UniqueRelevance," arXiv preprint, arXiv: 2212.06143, 2022.

[31] D. Choudhary and J.Shukla, "Feature Extraction and Feature Selection for Emotion Recognition UsingFacial Expression," Proc. of the 2020 IEEE 6th Int. Conf. on Multimedia Big Data (BigMM), pp. 125-133, New Delhi, India, September 2020.

[32] Z. Mortezaie and H. Hassanpour, "A Survey on Age-invariant Face Recognition Methods", JordanianJournal of Computers and Information Technology (JJCIT), vol. 05, no. 02, pp. 87 - 96, August 2019.

[33] M. Jiang and S. Yin, "Facial Expression Recognition Based on Convolutional Block Attention Moduleand Multi-feature Fusion," International Journal of Computational Vision and Robotics, vol. 13, no. 1, pp. 21-37, 2023.

[34] Y. Jiang and S. Yin, "Heterogenous-View Occluded Expression Data Recognition Based on Cycle- consistent Adversarial Network and K-SVD Dictionary Learning under Intelligent Cooperative Robot Environment," Computer Science and Information Systems, vol. 20, no. 4, pp. 1869-1883, 2023.

[35] X. Meng, X. Wang, S. Yin and H. Li, "Few-shot Image Classification Algorithm Based on AttentionMechanism and Weight Fusion," Journal of Engineering and Applied Science, vol. 70, no. 1, p. 14, 2023.

[36] L. Teng, "Brief Review of Medical Image Segmentation Based on Deep Learning," IJLAI Transactionson Science and Engineering, vol. 1, no. 02, pp. 01-08, 2023.

[37] N. El Aboudi and L. Benhlima, "Review on Wrapper Feature Selection Approaches," Proc. of the 2016IEEE Int. Conf. on Engineering & MIS (ICEMIS), pp. 1-5, Agadir, Morocco, September 2016.

[38] N. Aifanti, C. Papachristou and A. Delopoulos, "The MUG Facial Expression Database," Proc. of the11th IEEE Int. Workshop on Image Analysis for Multimedia Interactive Services (WIAMIS 10), pp. 1-4, Desenzano del Garda, Italy, April 2010.

[39] R. J. N. J. Bellman, Dynamic Programming, ISBN 0-691-07951-X, Princeton University Press,Princeton, New Jersey, 1957.

[40] I. Kononenko, E. Šimec and M. Robnik-Šikonja, "Overcoming the Myopia of Inductive LearningAlgorithms with RELIEFF," Applied Intelligence, vol. 7, pp. 39-55, 1997.

[41] H. Peng, F. Long and C. Ding, "Feature Selection Based on Mutual Information: Criteria of Max- dependency, Max-relevance and Min-redundancy," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 8, p. 1226-1238, 2005.

[42] H. Liu and R. Setiono, "Chi2: Feature Selection and Discretization of Numeric Attributes," Proc. of the7th IEEE Int. Conf. on Tools with Artificial Intelligence, pp. 388-391, Herndon, VA, USA, 1995.

[43] S. Mirjalili, "SCA: A Sine Cosine Algorithm for Solving Optimization Problems," Knowledge-basedSystems, vol. 96, pp.120-133, 2016.

[44] G. E. Hinton and R. R. Salakhutdinov, "Reducing the Dimensionality of Data with Neural Networks,"Science, vol. 313, no. 5786, pp. 504-507, 2006.

,abstract={Automatic Facial Expression Recognition (FER) systems provide an important way to express and interpret the emotional and mental states of human beings. These FER systems transform the facial image into a set of features to train a classifier capable of distinguishing between different classes of emotions. However, the problem often posed is that the extracted feature vectors sometimes contain irrelevant or redundant features, which decreases the accuracy of the induced classifier and increases the computation time. To overcome this problem, dimensionality must be reduced by selecting only the most relevant features. In this paper, we study the impact of adding the "Wrapper" selection approach and using the information provided by different local regions of the face such as the mouth, eyes and eyebrows, on the performance of a traditional FER system based on a local geometric feature-extraction method. The objective here is to test and analyze how this combination can improve the overall performance of the original traditional system. The obtained results, based on the Multimedia Understanding Group (MUG) database, showed that the FER system combined with the proposed feature-selection strategy gives better classification results than the original system for all four classification models; namely, K-Nearest Neighbor (KNN) classifier, Tree classifier, NB classifier and Linear Discriminant Analysis (LDA). Indeed, a considerable reduction (up to 50%) in the number of features used and an accuracy of 100%, using the LDA classifier, were observed, which represents a significant improvement in terms of computation time, efficiency and memory space. Furthermore, the majority of relevant features used are part of the "eyebrows’ region", which proves the importance of using information from local regions of the face in emotion recognition tasks.},
keywords={Facial expression recognition,Local feature extraction,Feature selection,Wrapper approach,Classification},
ISSN={2413-9351},
month={}}