Wireless Sensing Technology Combined with Facial Expression to Realize Multimodal Emotion Recognition
Abstract
:1. Introduction
- In this paper, we use millimeter-wave radar to capture heartbeat and respiration signals in different emotional states while combining three modal data of facial expression images, and then perform parallel fusion after feature extraction by respective CNN deep learning models and use the fused features as the input of GRU deep learning model, and then achieve the classification task of four emotions.
- Considering that the breathing and heartbeat signals are non-stationary signals and the noise problem of the experiment scene, the method of combining MTI and VMD is proposed, and the comparison experiment with an MI5 smartwatch proves that it can get high-quality signals.
- To prove the effectiveness and superiority of the proposed method, this paper not only sets up comparison tests with traditional machine learning methods and single deep learning models but also a comprehensive comparison from robustness and other advanced methods. The model is proven to have a more excellent classification ability through many experiments.
2. Relate Work
3. The Proposed Method
3.1. Working Principle of FMCW Radar
3.2. Data Preprocessing
3.2.1. MTI Removes Static Clutter
3.2.2. VMD Algorithm
3.2.3. Video Signal Preprocessing
3.3. Proposed Deep Learning Model
3.3.1. Construction of 1D-CNN
3.3.2. Construction of 2D-CNN
3.3.3. Construction of GRU
4. Experiments and Results
4.1. Experimental Design
4.2. Evaluating-Indicator and Cross Validation
4.3. Experimental Analysis
4.3.1. Emotion Recognition Accuracy of ER-MiCG
4.3.2. Robustness Testing
4.3.3. Comparison of Different Emotion Recognition Methods
4.3.4. Comparison with Traditional Machine Learning Algorithms
4.3.5. Comparison of Different Deep Learning Models
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Abbreviations
CSI | Channel State Information |
MTI | Moving Target Inidcation |
VMD | Variational Modal Decomposition |
1D-CNN | One-Dimensional Convolution Neural Network |
2D-CNN | Two-Dimensional Convolution Neural Network |
GRU | Gate Recurrent Unit |
ML | Machine Learning |
DL | Deep Learning |
FMCW | Frequency Modulated Continuous Wave |
CW | Continuous Wave |
UWB | Ultra-WideBand |
ELU | Exponential Linear Units |
LOOCV | Leave-One-Out Cross-Validation |
MSE | Mean Squared Error |
SVM | Support Vector Machine |
KNN | K-Nearest Neighbor |
RF | Random Forest |
FER | Face Emotion Recognition |
References
- Jazaieri, H.; Morrison, A.S.; Goldin, P.R.; Gross, J.J. The Role of Emotion and Emotion Regulation in Social Anxiety Disorder. Curr. Psychiatry Rep. 2015, 17, 531. [Google Scholar] [CrossRef] [PubMed]
- Rosenkranz, M.A.; Jackson, D.C.; Dalton, K.M.; Dolski, I.; Ryff, C.D.; Singer, B.H.; Muller, D.; Kalin, N.H.; Davidson, R.J. Affective style and in vivo immune response: Neurobehavioral mechanisms. Proc. Natl. Acad. Sci. USA 2003, 100, 11148–11152. [Google Scholar] [CrossRef] [Green Version]
- Hittner, E.F.; Stephens, J.E.; Turiano, N.A.; Gerstorf, D.; Lachman, M.E.; Haase, C.M. Positive Affect Is Associated with Less Memory Decline: Evidence from a 9-Year Longitudinal Study. Psychol. Sci. 2020, 31, 1386–1395. [Google Scholar] [CrossRef] [PubMed]
- Sadikin, F.; Deursen, T.V.; Kumar, S. A ZigBee Intrusion Detection System for IoT using Secure and Efficient Data Collection. Internet Things 2020, 12, 100306. [Google Scholar] [CrossRef]
- Tan, S.; Yang, J.; Chen, Y.Y. Enabling Fine-Grained Finger Gesture Recognition on Commodity WiFi Devices. IEEE Trans. Mob. Comput. 2022, 21, 2789–2802. [Google Scholar] [CrossRef]
- He, M.; Nian, Y.J.; Gong, Y.S. Novel signal processing method for vital sign monitoring using FMCW radar. Biomed. Signal Process. Control 2017, 33, 335–345. [Google Scholar] [CrossRef]
- Feng, D.Q.; Wang, C.Q.; He, C.L.; Zhuang, Y.; Xia, X.G. Kalman-Filter-Based Integration of IMU and UWB for High-Accuracy Indoor Positioning and Navigation. IEEE Internet Things J. 2020, 7, 3133–3146. [Google Scholar] [CrossRef]
- Anghel, A.; Vasile, G.; Cacoveanu, R.; Ioana, C.; Ciochina, S. Short-Range Wideband FMCW Radar for Millimetric Displacement Measurements. IEEE Trans. Geosci. Remote Sens. 2014, 52, 5633–5642. [Google Scholar] [CrossRef] [Green Version]
- Qu, K.; Zhang, R.F.; Fang, Z.J. High Accuracy Motion Detection Algorithm via ISM Band FMCW Radar. Remote Sens. 2022, 14, 58. [Google Scholar] [CrossRef]
- Shen, X.Y.; Zheng, H.F.; Feng, X.X.; Hu, J.S. ML-HGR-Net: A Meta-Learning Network for FMCW Radar Based Hand Gesture Recognition. IEEE Sens. J. 2022, 22, 10808–10817. [Google Scholar] [CrossRef]
- Alarcão, S.; Fonseca, M.J. Emotions Recognition Using EEG Signals: A Survey. IEEE Trans. Affect. Comput. 2017, 10, 374–393. [Google Scholar] [CrossRef]
- Woodworth, R.S. Principles of physiological psychology. Science 1905, 22, 789–790. [Google Scholar] [CrossRef] [Green Version]
- Russell, J.A. A circumplex model of affect. J. Personal. Soc. Psychol. 1980, 39, 1161. [Google Scholar] [CrossRef]
- Posner, J.; Russell, J.A.; Peterson, B.S. The circumplex model of affect: An integrative approach to affective neuroscience, cognitive development, and psychopathology. Dev. Psychopathol. 2005, 17, 715–734. [Google Scholar] [CrossRef] [PubMed]
- Mri, C.; Mmi, B.; Mmr, D.; Cm, A.; Sks, A.; Ma, A.; Maa, F.; Msi, E.; Mam, G. EEG Channel Correlation Based Model for Emotion Recognition. Comput. Biol. Med. 2021, 136, 104757. [Google Scholar]
- Islam, M.R.; Moni, M.A.; Islam, M.M.; Rashed-Al-Mahfuz, M.; Lio, P. Emotion Recognition from EEG Signal Focusing on Deep Learning and Shallow Learning Techniques. IEEE Access 2021, 9, 94601–94624. [Google Scholar] [CrossRef]
- Wang, S.Y. Online Learning Behavior Analysis Based on Image Emotion Recognition. Trait. Signal 2021, 38, 865–873. [Google Scholar] [CrossRef]
- Neves, L.; Martins, M.; Correia, A.I.; Castro, S.L.; Lima, C.F. Associations between vocal emotion recognition and socio-emotional adjustment in children. R. Soc. Open Sci. 2021, 8, 16. [Google Scholar] [CrossRef]
- Canal, F.Z.; Muller, T.R.; Matias, J.C.; Scotton, G.G.; De Sjar, A.R.; Pozzebon, E.; Sobieranski, A.C. A survey on facial emotion recognition techniques: A state-of-the-art literature review. Inf. Sci. 2022, 582, 593–617. [Google Scholar] [CrossRef]
- Hasnul, M.A.; Aziz, N.A.A.A.; Alelyani, S.; Mohana, M.; Aziz, A.A.A. Electrocardiogram-Based Emotion Recognition Systems and Their Applications in Healthcare—A Review. Sensors 2021, 21, 37. [Google Scholar] [CrossRef]
- Yang, L.Y.; Qin, S.F. A Review of Emotion Recognition Methods From Keystroke, Mouse, and Touchscreen Dynamics. IEEE Access 2021, 9, 162197–162213. [Google Scholar] [CrossRef]
- Kulkarni, K.; Corneanu, C.; Ofodile, I.; Escalera, S.; Baro, X.; Hyniewska, S.; Allik, J.; Anbarjafari, G. Automatic Recognition of Facial Displays of Unfelt Emotions. IEEE Trans. Affect. Comput. 2021, 12, 377–390. [Google Scholar] [CrossRef] [Green Version]
- Li, Y.K.; Meng, Q.H.; Yang, T.H.; Wang, Y.X.; Hou, H.R. Touch Gesture and Emotion Recognition Using Decomposed Spatiotemporal Convolutions. IEEE Trans. Instrum. Meas. 2022, 71, 9. [Google Scholar] [CrossRef]
- Sheng, W.J.; Li, X.D. Multi-task learning for gait-based identity recognition and emotion recognition using attention enhanced temporal graph convolutional network. Pattern Recognit. 2021, 114, 12. [Google Scholar] [CrossRef]
- Fan, W.Q.; Xu, X.M.; Cai, B.L.; Xing, X.F. ISNet: Individual Standardization Network for Speech Emotion Recognition. IEEE-ACM Trans. Audio Speech Lang. Process. 2022, 30, 1803–1814. [Google Scholar] [CrossRef]
- Shukla, J.; Barreda-Angeles, M.; Oliver, J.; Nandi, G.C.; Puig, D. Feature Extraction and Selection for Emotion Recognition from Electrodermal Activity. IEEE Trans. Affect. Comput. 2021, 12, 857–869. [Google Scholar] [CrossRef]
- Jenke, R.; Peer, A.; Buss, M. Feature Extraction and Selection for Emotion Recognition from EEG. IEEE Trans. Affect. Comput. 2014, 5, 327–339. [Google Scholar] [CrossRef]
- Jz, A.; Zhong, Y.B.; Peng, C.C.; Sn, A. Emotion recognition using multi-modal data and machine learning techniques: A tutorial and review. Inf. Fusion 2020, 59, 103–126. [Google Scholar]
- Sheikhan, M.; Bejani, M.; Gharavian, D. Modular neural-SVM scheme for speech emotion recognition using ANOVA feature selection method. Neural Comput. Appl. 2013, 23, 215–227. [Google Scholar] [CrossRef]
- Tiwari, P.; Rathod, H.; Thakkar, S.; Darji, A.D. Multimodal emotion recognition using SDA-LDA algorithm in video clips. J. Ambient. Intell. Humaniz. Comput. 2021. [Google Scholar] [CrossRef]
- Gao, Z.; Wang, X.; Yang, Y.; Li, Y.; Ma, K.; Chen, G. A Channel-fused Dense Convolutional Network for EEG-based Emotion Recognition. IEEE Trans. Cogn. Dev. Syst. 2020, 13, 945–954. [Google Scholar] [CrossRef]
- Khalil, R.A.; Jones, E.; Babar, M.I.; Jan, T.; Alhussain, T. Speech Emotion Recognition Using Deep Learning Techniques: A Review. IEEE Access 2019, 7, 117327–117345. [Google Scholar] [CrossRef]
- Kwon, S. MLT-DNet: Speech emotion recognition using 1D dilated CNN based on multi-learning trick approach. Expert Syst. Appl. 2020, 167, 114177. [Google Scholar]
- Sreenivas, V.; Namdeo, V.; Kumar, E.V. Modified deep belief network based human emotion recognition with multiscale features from video sequences. Softw.-Pract. Exp. 2021, 51, 1259–1279. [Google Scholar] [CrossRef]
- Wang, G.; Gu, C.; Inoue, T.; Li, C. A Hybrid FMCW-Interferometry Radar for Indoor Precise Positioning and Versatile Life Activity Monitoring. IEEE Trans. Microw. Theory Tech. 2014, 62, 2812–2822. [Google Scholar] [CrossRef]
- Zhang, Z.Y.; Tian, Z.S.; Zhou, M. Latern: Dynamic Continuous Hand Gesture Recognition Using FMCW Radar Sensor. IEEE Sens. J. 2018, 18, 3278–3289. [Google Scholar] [CrossRef]
- Li, H.; Shrestha, A.; Heidari, H.; Le Kernec, J.; Fioranelli, F. Bi-LSTM Network for Multimodal Continuous Human Activity Recognition and Fall Detection. IEEE Sens. J. 2020, 20, 1191–1201. [Google Scholar] [CrossRef] [Green Version]
- Alizadeh, M.; Shaker, G.; Almeida, J.D.; Morita, P.P.; Safavi-Naeini, S. Remote Monitoring of Human Vital Signs Using mm-Wave FMCW Radar. IEEE Access 2019, 7, 54958–54968. [Google Scholar] [CrossRef]
- Peng, Z.; Muñoz-Ferreras, J.M.; Tang, Y.; Liu, C.; Gómez-García, R.; Ran, L.; Li, C. A Portable FMCW Interferometry Radar with Programmable Low-IF Architecture for Localization, ISAR Imaging, and Vital Sign Tracking. IEEE Trans. Microw. Theory Tech. 2017, 65, 1334–1344. [Google Scholar] [CrossRef]
- Dang, X.C.; Chen, Z.T.; Hao, Z.J. Emotion recognition method using millimetre wave radar based on deep learning. IET Radar Sonar Navig. 2022, 16, 1796–1808. [Google Scholar] [CrossRef]
- Dragomiretskiy, K.; Zosso, D. Variational Mode Decomposition. IEEE Trans. Signal Process. 2014, 62, 531–544. [Google Scholar] [CrossRef]
- Sanchez-Reolid, R.; Francisco, L.; Lopez, M.T.; Fernandez-Caballero, A. One-dimensional convolutional neural networks for low/high arousal classification from electrodermal activity. Biomed. Signal Process. Control 2022, 71, 9. [Google Scholar] [CrossRef]
- Zhao, M.; Adib, F.; Katabi, D. Emotion Recognition Using Wireless Signals. Commun. ACM 2018, 61, 91–100. [Google Scholar] [CrossRef]
- Gu, Y.; Wang, Y.; Liu, T.; Ji, Y.; Liu, Z.; Li, P.; Wang, X.; An, X.; Ren, F. EmoSense: Computational Intelligence Driven Emotion Sensing via Wireless Channel Data. IEEE Trans. Emerg. Top. Comput. Intell. 2020, 4, 216–226. [Google Scholar] [CrossRef] [Green Version]
- Khan, A.N.; Ihalage, A.A.; Ma, Y.; Liu, B.; Liu, Y.; Hao, Y. Deep learning framework for subject-independent emotion detection using wireless signals. PLoS ONE 2021, 16, 16. [Google Scholar] [CrossRef]
- Jain, N.; Kumar, S.; Kumar, A.; Shamsolmoali, P.; Zareapoor, M. Hybrid deep neural networks for face emotion recognition. Pattern Recognit. Lett. 2018, 115, 101–106. [Google Scholar] [CrossRef]
Types | Kernel Size | No. of Filters | Stride |
---|---|---|---|
1D Full convolution | 5 | 16 | 1 |
Max pooling | 4 | 16 | 4 |
1D Full convolution | 5 | 32 | 1 |
Max pooling | 2 | 32 | 2 |
1D Full convolution | 5 | 32 | 2 |
Max pooling | 2 | 32 | 2 |
1D Full convolution | 5 | 64 | 2 |
Max pooling | 2 | 64 | 2 |
Types | Kernel Size | No. of Filters | Stride |
---|---|---|---|
2D Full convolution | 3 × 3 | 32 | 1 |
Max pooling | 2 × 2 | 32 | 2 |
2D Full convolution | 3 × 3 | 64 | 1 |
Max pooling | 2 × 2 | 64 | 2 |
2D Full convolution | 3 × 3 | 96 | 1 |
Max pooling | 4 × 4 | 96 | 4 |
2D Full convolution | 3 × 3 | 96 | 1 |
Max pooling | 4 × 4 | 96 | 4 |
Parameters | Kernel Size |
---|---|
Number of Transmitting Antennas | 1 |
Number of Receiving Antennas | 4 |
Carrier Frequency | 77 GHz |
Bandwidth | 4 GHz |
frequency modulation | 66.62 MHz/μs |
Single Chirp Signal Duration | 60 ηs |
Period of frame | 50 ms |
Number of Chirps per Frame | 128 |
Number of Frames | 150 |
Number of Samples per Chirp | 256 |
Project | Emotoion | Algorithm | Feature | Average Accuracy |
---|---|---|---|---|
ER-MICG | Relax, Happy, Sad, Anger | CNN and GRU | FMCW and FER | 74.25% |
EQ-Radio [43] | Joy, Pleasure, Sad, Anger | SVM | FMCW | 72.3% |
EmoSense [44] | Happy, Sad, Anger, Fear | KNN | CSI | 40.86% |
Yang Hao et al. [45] | Scary, Relax, Joy, Disgust | CNN and LSTM | FMCW and Continuous wavelet images | 71.67% |
Jain, N., et al. [46] | Fear, Happy, Sad, Anger, Surprise, Disgust, Neutral | CNN and RNN | FER | 93.49% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Dang, X.; Chen, Z.; Hao, Z.; Ga, M.; Han, X.; Zhang, X.; Yang, J. Wireless Sensing Technology Combined with Facial Expression to Realize Multimodal Emotion Recognition. Sensors 2023, 23, 338. https://doi.org/10.3390/s23010338
Dang X, Chen Z, Hao Z, Ga M, Han X, Zhang X, Yang J. Wireless Sensing Technology Combined with Facial Expression to Realize Multimodal Emotion Recognition. Sensors. 2023; 23(1):338. https://doi.org/10.3390/s23010338
Chicago/Turabian StyleDang, Xiaochao, Zetong Chen, Zhanjun Hao, Macidan Ga, Xinyu Han, Xiaotong Zhang, and Jie Yang. 2023. "Wireless Sensing Technology Combined with Facial Expression to Realize Multimodal Emotion Recognition" Sensors 23, no. 1: 338. https://doi.org/10.3390/s23010338
APA StyleDang, X., Chen, Z., Hao, Z., Ga, M., Han, X., Zhang, X., & Yang, J. (2023). Wireless Sensing Technology Combined with Facial Expression to Realize Multimodal Emotion Recognition. Sensors, 23(1), 338. https://doi.org/10.3390/s23010338