• Dr. Deepthi V S
• Dr. Vagdevi S

The burgeoning interest in harnessing electroencephalography (EEG) signals for non-muscular communication and control has spurred extensive research in pattern recognition. Recent years have witnessed a surge in efforts to extract meaningful features from EEG data, aiming to elucidate the intricate connections between brain activity and behavior. However, conventional vectorization-based feature representations, be it vector-like or matrix-like, are hampered by pervasive signal noise and the challenge of leveraging signal correlations among neighboring EEG sensors. A crucial preliminary step in integrating EEG signals into a learning model entails standardizing them into a unified frequency representation. Disregarding specific frequency components of EEG signals can undermine activity recognition, as different frequencies hold varying degrees of relevance in discerning distinct activities. The proposed approach hinges on three fundamental pillars: data preprocessing, feature extraction, and model training. Data preparation encompasses artifact avoidance and linear filtering to mitigate noise interference. Notably, the Common Spatial Domain emerges as the predominant choice for feature extraction. Subsequently, the Hybrid EEGNET model is deployed for model training, demonstrating superior performance compared to the commonly employed alternatives, namely CNN and LSTM. Remarkably, the proposed method achieves an impressive success rate of 97.52 percent.

Electroencephalography (EEG), Inertial Measurement Unit (IMU), Motion intention recognition.

"Unveiling Human Intentions: EEGNET's Hybrid Role in Motion Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 5, page no.1-5, May-2024, Available :http://www.jetir.org/papers/JETIRGG06001.pdf

"Unveiling Human Intentions: EEGNET's Hybrid Role in Motion Detection", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 5, page no. pp1-5, May-2024, Available at : http://www.jetir.org/papers/JETIRGG06001.pdf