Emotion Recognition from DEAP: Feature Engineering & Baselines
Published:
Preprint codebase for DEAP EEG emotion recognition with multiple handcrafted features and CNN models. 
Posts by Collection
portfolio
Smart Classroom Attendance & Automation System
Published:
Prototype smart classroom system combining voice, gesture, and face recognition for automated attendance and control.
Real-World EEG Emotion Recognition (SEED Pilot)
Published:
Pilot implementation for real-world EEG emotion recognition using compact deep models on SEED-like setups.
Time–Frequency EEG Emotion Recognition on DEAP & SEED
Published:
End-to-end pipeline for time–frequency feature extraction and 3D-CNN based emotion recognition on DEAP and SEED EEG datasets.
Equivariant Geodesic Networks (EGN)
Published:
Library of geometry-preserving layers and losses for learning on Riemannian manifolds, including SPD-valued features.
publications
Recognition of Bengali Vowels from Auditory Evoked Potentials Using CNN
Published in IEEE International Conference on Signal Processing, Information, Communication and Systems (SPICSCON), 2024, 2024
A CNN-based approach for recognizing Bengali vowels from auditory evoked potential (AEP) EEG signals.
Recommended citation: Tithi Das, Md Raihan Khan, and Md Mahbub Hasan, "Recognition of Bengali Vowels from Auditory Evoked Potentials Using CNN," IEEE SPICSCON 2024.
Download Paper
Machine Learning Techniques for Brain Stroke Analysis and Prediction
Published in IEEE International Conference on Signal Processing, Information, Communication and Systems (SPICSCON), 2024, 2024
A comparative study of machine learning techniques for brain stroke analysis and risk prediction.
Recommended citation: Rabita Hasan, Sheikh Md Rabiul Islam, and Md Raihan Khan, "Machine Learning Techniques for Brain Stroke Analysis and Prediction," IEEE SPICSCON 2024.
Download Paper
Mental Stress Detection from EEG Signals Using Comparative Analysis of Random Forest and Recurrent Neural Network
Published in IEEE International Conference on Advances in Computing, Communication, Electrical, and Smart Systems (iCACCESS), 2024, 2024
This study compares Random Forest and RNN-based models for mental stress detection from EEG signals.
Recommended citation: Md Raihan Khan and Mohiuddin Ahmad, "Mental Stress Detection from EEG Signals Using Comparative Analysis of Random Forest and Recurrent Neural Network," IEEE iCACCESS 2024.
Download Paper
Smart Classroom Automation: A Fusion of AI with Voice, Gesture, and Face Recognition Attendance System
Published in IEEE International Conference on Advances in Computing, Communication, Electrical, and Smart Systems (iCACCESS), 2024, 2024
An AI-driven smart classroom system integrating voice, gesture, and face recognition for automated attendance and interaction.
Recommended citation: Md Raihan Khan, Abdullah Al Ahad, Airin Akter Tania, Tithi Das, and Bibhakar Das, "Smart Classroom Automation: A Fusion of AI with Voice, Gesture, and Face Recognition Attendance System," IEEE iCACCESS 2024.
Download Paper
Multiclass Liver Disease Prediction with Adaptive Data Preprocessing and Ensemble Modeling
Published in Results in Engineering, Elsevier, 2025, 2024
This work proposes an adaptive preprocessing pipeline and ensemble modeling framework for multiclass liver disease prediction on the Hepatitis C dataset, achieving up to 99.80% test accuracy.
Recommended citation: Abdullah Al Ahad, Bibhakar Das, Md Raihan Khan, Nitol Saha, Abu Zahid, and Mohiuddin Ahmad, "Multiclass liver disease prediction with adaptive data preprocessing and ensemble modeling," Results in Engineering, Elsevier, 2025.
Download Paper
Custom Dataset-driven Unsupervised Low-light Image Enhancement using 2D CNN
Published in IEEE International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN), 2025, 2025
An unsupervised 2D CNN-based approach for low-light image enhancement on a custom dataset.
Recommended citation: Airin Akter Tania, Md Raihan Khan, and Mohiuddin Ahmad, "Custom Dataset-driven Unsupervised Low-light Image Enhancement using 2D CNN," IEEE QPAIN 2025.
Download Paper
A Comparative Study of Time–Frequency Features Based Spatio-temporal Analysis with Varying Multiscale Kernels for Emotion Recognition from EEG
Published in Biomedical Signal Processing and Control, Elsevier, 2025, 2025
This work compares multiple time–frequency features and multiscale 3D CNN kernels for EEG-based emotion recognition using DEAP and SEED.
Recommended citation: Md Raihan Khan, Airin Akter Tania, and Mohiuddin Ahmad, "A comparative study of time–frequency features based spatio-temporal analysis with varying multiscale kernels for emotion recognition from EEG," Biomedical Signal Processing and Control, Elsevier, 2025.
Download Paper
DAFNet: A Dual-path Attention Fusion Network for EEG Emotion Recognition via CNN and Graph-based Global Modeling
Published in Array, Elsevier (Available online 4 November 2025), 2025
DAFNet is a dual-path attention fusion network that combines local CNN-based processing of differential entropy features with graph-based modeling of spectral coherence symmetry for robust EEG emotion recognition.
Recommended citation: Md Raihan Khan, Airin Akter Tania, Tanjum Arifen Bushra, Jahanara Pritha, and Mohiuddin Ahmad, "DAFNet: A dual-path attention fusion network for EEG emotion recognition via CNN and graph-based global modeling," Array, Elsevier, 2025 (Available online 4 November 2025).
Download Paper
AutoLumNet: A Bi-Branch Exposure-Aware Network for Low- and High-Exposure Image Enhancement
Published in International Conference on Learning Representations (ICLR), 2026 – Under Review, 2026
AutoLumNet is a bi-branch exposure-aware network that jointly enhances low- and high-exposure images using complementary correction paths.
Recommended citation: Airin Akter Tania, Md Raihan Khan, and Mohiuddin Ahmad, "AutoLumNet: A Bi-Branch Exposure-Aware Network for Low- and High-Exposure Image Enhancement," ICLR 2026 (under review).
Download Paper
Equivariant Geodesic Networks: Geometry Preserving Learning on Riemannian Manifolds
Published in International Conference on Learning Representations (ICLR), 2026 – Under Review, 2026
Equivariant Geodesic Networks (EGN) provide a geometry-preserving framework for learning on Riemannian manifolds, with applications to covariance-based representations.
Recommended citation: Md Raihan Khan and Airin Akter Tania, "Equivariant Geodesic Networks: Geometry Preserving Learning on Riemannian Manifolds," ICLR 2026 (under review).
Download Paper
Searching for the Best Polynomial Approximation for the Accurate Log Matrix Normalization in Global Covariance Pooling
Published in International Conference on Learning Representations (ICLR), 2026 – Under Review, 2026
This work systematically studies polynomial approximations of the matrix logarithm in Global Covariance Pooling, targeting accurate but efficient log-normalization.
Recommended citation: Md Rifat Ur Rahman and Md Raihan Khan, "Searching for the Best Polynomial Approximation for the Accurate Log Matrix Normalization in Global Covariance Pooling," ICLR 2026 (under review).
Download Paper
talks
teaching
Lecturer, Department of Electrical and Electronic Engineering
University teaching, North Western University, Department of EEE, 2024
Since August 2024, I have been serving as a Lecturer in the Department of Electrical and Electronic Engineering at North Western University, Bangladesh. My teaching responsibilities include delivering lectures, conducting laboratory sessions, designing assignments, and supervising student projects in the following courses: