Automated Detection of Micro-Expressions Using Time and Frequency Domain Features of Facial Electromyography Gaadha Jeevan, Gobinath Kaliyaperumal, Mythili A, Santhi R, Karthick P A Proceedings of 2024 International Conference on Brain Computer Interface and Healthcare Technologies Icon Bciht 2024, 2024 Facial Electromyography plays crucial role in the field of psychological evaluations by employing it in emotion detection and Micro-Expression Recognition (MER) due to its ability to detect facial muscle activity. The non-invasive behaviour and inter-subject variations can be leveraged by psychologists for psychotherapy and deception detection during therapy sessions. Micro-expressions (MEs) are low intensity and involuntary facial expressions that said to reveal true emotions which are often suppressed. In this study, an attempt has been made to detect and recognize Micro-Expression using Facial Electromyography. In this study, four emotion classes: - Happy, Sad, Fear and Anger are considered. The target muscle regions include Mentalis Region, Zygomaticus Region, Frontalis Region, and Orbital Region. The participants were presented with emotional videos, while voluntarily suppressing emotions with Facial EMG (fEMG) obtained simultaneously. The data was pre-processed and segmented based on Action Unit (AU) Coding. Temporal-domain features such as Mean Absolute Value (MAV), Root Mean Square (RMS), Variance (V AR), Integrated EMG (iEMG)and Zero Crossing Factor (ZCF) are extracted. Additionally, Spatial- domain features such as Mean Frequency (MF), Median Frequency (MDF) and Total Power (TP) are obtained. The classification of Micro-expressions is performed using Random Forest algorithm. This approach resulted in a mean Fl-score of 0.8525 and these insights play an essential role in decoding suppressed emotions.
Investigation on the microstructure, microhardness, and tribological behavior of AA1100-hBN surface composite R. Premkumar, R. V. Vignesh, R. Padmanaban, M. Govindaraju, R. Santhi Koroze A Ochrana Materialu, 2021 Aluminum alloy AA1100 has less wear resistance and mechanical properties than that of other aluminum alloys. This research work is on the fabrication of surface composites of AA1100 alloy by friction stir processing (FSP). The surface composites are fabricated by reinforcing hBN (hexagonal Boron Nitride) in AA1100 alloy to improve the mechanical and tribological properties. The influence of process parameters, rotational speed (rpm), and transverse speed (mm/min) on the microstructural evolution and properties of the fabricated surface composites is investigated.
