Facial Emotion Classification Using Transfer Learning with NASNetMobile on the AffectNet Dataset
Article Sidebar
Main Article Content
Abstract
Facial expressions represent one of the most important forms of nonverbal communication for conveying human emotions. However, automatic facial emotion classification remains a challenging task due to variations in illumination, facial pose, background complexity, and individual facial characteristics. This study aims to analyze facial emotion classification using a deep learning approach based on Convolutional Neural Networks (CNNs). The experiments were conducted on the AffectNet dataset, a large-scale benchmark containing in-the-wild facial images, with the classification task focused on two emotional categories: happy and fear. The proposed framework consists of several preprocessing stages, including face detection, image resizing, normalization, and data augmentation, followed by CNN training using a transfer learning approach with NASNetMobile as the feature extraction backbone. Experimental results demonstrate that the proposed model achieved a test accuracy of 96%, with F1-scores of 0.97 for the happy class and 0.95 for the fear class. The model exhibited superior performance in recognizing the happy emotion, whereas the fear class produced relatively more misclassifications due to its greater visual complexity and higher intra-class variability. Overall, the findings indicate that the combination of appropriate image preprocessing, data augmentation, and transfer learning significantly enhances the performance and generalization capability of CNN-based facial emotion classification under real-world conditions.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
How to Cite
References
[1]. A. K. S. Canal, T. A. S. R. S. M. G. P. G., and A. L. C. M. S. A. S., “A survey on facial emotion recognition techniques: A state-of-the-art literature review,” *Information Sciences*, vol. 582, pp. 593–617, 2022, doi: 10.1016/j.ins.2021.10.005.
[2]. M. Karnati, A. Seal, D. Bhattacharjee, A. Yazidi, and O. Krejcar, “Understanding deep learning techniques for recognition of human emotions using facial expressions: A comprehensive survey,” *IEEE Transactions on Instrumentation and Measurement*, vol. 72, pp. 1–31, 2023, doi: 10.1109/TIM.2023.3243661.
[3]. A. R. Khan, “Facial emotion recognition using conventional machine learning and deep learning methods: Current achievements, analysis and remaining challenges,” *Information*, vol. 13, no. 6, Art. no. 268, 2022, doi: 10.3390/info13060268.
[4]. N. Rathour, R. Singh, A. Gehlot, S. V. Akram, A. K. Thakur, and A. Kumar, “The decadal perspective of facial emotion processing and recognition: A survey,” *Displays*, vol. 75, Art. no. 102330, 2022, doi: 10.1016/j.displa.2022.102330.
[5]. R. Kaur, S. S. Bhatia, and others, “Facial emotion recognition: A comprehensive review,” *Expert Systems*, 2024, doi: 10.1111/exsy.13670.
[6]. N. N. Ali and A. M. Abdulazeez, “Facial emotion recognition based on deep learning: A review,” *International Journal of Research and Applied Technology (INJURATECH)*, vol. 4, no. 1, pp. 21–34, 2024.
[7]. M. C. Gursesli, L. Sara, M. Duradoni, L. Bocchi, A. Guazzini, and A. Lanata, “Facial Emotion Recognition (FER) Through Custom Lightweight CNN Model: Performance Evaluation in Public Datasets,” *IEEE Access*, vol. 12, pp. 45543–45559, 2024, doi: 10.1109/ACCESS.2024.3380847.
[8]. I. Na, A. Aldrees, A. Hakeem, L. Mohaisen, M. Umer, D. A. AlHammadi, S. Alsubai, N. Innab, and I. Ashraf, “FacialNet: Facial emotion recognition for mental health analysis using UNet segmentation with transfer learning model,” *Frontiers in Computational Neuroscience*, vol. 18, Art. no. 1485121, 2024, doi: 10.3389/fncom.2024.1485121.
[9]. S. Porcu, A. Floris, and L. Atzori, “Evaluation of data augmentation techniques for facial expression recognition systems,” *Electronics*, vol. 9, no. 11, Art. no. 1892, 2020, doi: 10.3390/electronics9111892.
[10]. Z. Zhao, Q. Liu, and F. Zhou, “Robust lightweight facial expression recognition network with label distribution training,” in *Proc. AAAI Conf. Artificial Intelligence*, vol. 35, no. 4, pp. 3510–3519, 2021, doi: 10.1609/aaai.v35i4.16465.
[11]. J. R. Lee, L. Wang, and A. Wong, “EmotionNet Nano: An efficient deep convolutional neural network design for real-time facial expression recognition,” *Frontiers in Artificial Intelligence*, vol. 3, Art. no. 609673, 2021, doi: 10.3389/frai.2020.609673.
[12]. D. Gera and S. Balasubramanian, “Landmark guidance independent spatio-channel attention and complementary context information based facial expression recognition,” *Pattern Recognition Letters*, vol. 145, pp. 58–66, 2021, doi: 10.1016/j.patrec.2021.01.029.
[13]. S. Umer, R. K. Rout, C. Pero, and M. Nappi, “Facial expression recognition with trade-offs between data augmentation and deep learning features,” *Journal of Ambient Intelligence and Humanized Computing*, vol. 12, no. 2, pp. 721–735, 2021, doi: 10.1007/s12652-020-02845-8.
[14]. H. Siqueira, S. Magg, and S. Wermter, “Efficient facial feature learning with wide ensemble-based convolutional neural networks,” in *Proc. AAAI Conf. Artificial Intelligence*, vol. 34, no. 4, pp. 5800–5809, 2020, doi: 10.1609/aaai.v34i04.6037.
[15]. Y. Cai, J. Gao, G. Zhang, and Y. Liu, “Efficient facial expression recognition based on convolutional neural network,” *Intelligent Data Analysis*, vol. 25, no. 1, 2021, doi: 10.3233/IDA-194965.
[16]. W. G. Colares, M. G. F. Costa, and C. F. F. Costa Filho, “Enhancing emotion recognition: A dual-input model for facial expression recognition using images and facial landmarks,” in *Proc. IEEE Engineering in Medicine and Biology Society (EMBC)*, 2024, pp. 1–5, doi: 10.1109/EMBC53108.2024.10782924.
[17]. M. Mollahosseini, B. Hasani, and M. H. Mahoor, “AffectNet: A database for facial expression, valence, and arousal computation in the wild,” *IEEE Transactions on Affective Computing*, vol. 10, no. 1, pp. 18–31, 2019, doi: 10.1109/TAFFC.2017.2740923.
[18]. “Advances in facial expression recognition: A survey of methods, benchmarks, models, and datasets,” *Information*, vol. 15, no. 3, Art. no. 135, 2024, doi: 10.3390/info15030135.
[19]. M. I. Maulana, K. Aeni, and F. Fathulloh, “Pengenalan ekspresi wajah menggunakan Convolutional Neural Network (CNN),” *Indonesian Journal of Informatics and Research*, vol. 4, no. 2, 2023, doi: 10.58436/ijir.v4i2.1905.
[20]. A. S. Guntoro, E. Julianto, and D. Budiyanto, “Pengenalan ekspresi wajah menggunakan Convolutional Neural Network,” *Jurnal Informatika Atma Jogja*, vol. 3, no. 2, pp. 155–160, 2022, doi: 10.24002/jiaj.v3i2.6790.
[21]. S. Sutarti and F. Syaqialloh, “Klasifikasi dan pengenalan emosi dari ekspresi wajah menggunakan CNN-BiLSTM dengan teknik data augmentation,” *Decode: Jurnal Pendidikan Teknologi Informasi*, vol. 5, no. 1, 2025, doi: 10.51454/decode.v5i1.1038.