Authors:
R. Angeline, A. Alice Nithya

Addresses:
Department of Computational Intelligence, SRM Institute of Science and Technology, Kattankulathur, Chennai, Tamil Nadu, India.

Abstract:

The digital education revolution affords students and teachers worldwide unprecedented opportunities. Student involvement, comprehension, and academic integrity can be challenging in online classrooms. Online education requires more effective ways to track and enhance student engagement. Teachers evaluate student attention and understanding with images and brief exchanges in traditional classrooms. Student-specific instruction in online classes is impossible without these identifiers. Online students may show emotion through microexpressions. This rapid, involuntary expression shows students' immediate emotions and attentiveness. In movies and blurred images, CASME II's facial emotional recognition shows student interest. Microexpressions help teachers identify distracted students and adjust their teaching strategies. Real-time feedback helped teachers better understand their students and enhance online education. Understanding microexpressions can change schooling. A teacher who can detect students' mild dissatisfaction, bewilderment, or interest. They can instantly support kids and establish a compassionate learning environment. Cheating and academic dishonesty are detected to ensure fair assessments and the value of the assessment. Teachers track student improvement with CASME II's microexpression analysis dataset. Similar to a crystal ball that predicts educational success. CASME II detects micro-expressions 43.51% accurately. To increase the accuracy and multimodal facial expression and micro expression analysis for 5-class classification to 84% Image Feature Stabilisation Algorithm accuracy, so that every student can succeed.

Keywords: Student Engagement; Involuntary Expression; Micro Expression; Emotional Language; Online Education Grows; CASME II Dataset; Visuals and Brief Exchanges.

Received on: 01/10/2024, Revised on: 05/12/2024, Accepted on: 09/01/2025, Published on: 03/06/2025

DOI: 10.69888/FTSCL.2025.000424

FMDB Transactions on Sustainable Computer Letters, 2025 Vol. 3 No. 2, Pages: 85-96

• Views : 74
• Downloads : 9

Download PDF