An Efficient-attention-Enchanced CNN–BiLSTM Architecture for Automatic Emotion Recognition

Author(s)	Sriranjani P, Sowmiya K, Sujitha A, Udhayasri E, Saranya P
Country	India
Abstract	Facial emotion recognition (FER) continues to pose considerable difficulties owing to real-world constraints such as image degradation, inconsistent illumination, and severe class imbalance in benchmark datasets. To overcome these shortcomings, this work introduces a hybrid deep learning framework that unifies a Convolutional Neural Network (CNN) reinforced with Residual Network (ResNet) skip connections, a Bidirectional Long Short-Term Memory (Bi-LSTM) network, and a soft attention mechanism. A denoising autoencoder (DAE) is employed at the preprocessing stage to suppress noise and recover fine facial detail before feature extraction begins. Experiments conducted on the publicly available FER2013 benchmark confirm that the proposed model yields higher classification accuracy and greater resilience under adverse imaging conditions than conventional architectures, establishing its suitability for deployment in real-time affective computing scenarios.
Keywords	Facial Emotion Recognition, CNN–BiLSTM, Attention Mechanism, Deep Learning, FER2013
Field	Computer Applications
Published In	Volume 7, Issue 2, March-April 2026
Published On	2026-04-18

About AIJFR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9898-79-39-59	editor@aijfr.com

Advanced International Journal for Research