Main Article Content

S. Venkata Ramana, Ayesha Begum, P. Sindhu, M. Hepsheeba, M. Chandana


Facial expression recognition (FER) plays a crucial role in understanding human emotions and behaviors, particularly in the context of child development and attention assessment. This study presents a novel approach for detecting and classifying facial expressions in children to analyze their attentional states using Support Vector Machine (SVM) algorithm.

The proposed system utilizes a large dataset of facial images of children captured during various attention-demanding tasks. Preprocessing techniques are applied to extract relevant facial features, including key points and facial landmarks. A feature extraction process based on deep learning is employed to represent high-dimensional facial information in a compact manner. Next, the extracted features are fed into an SVM classifier to categorize facial expressions into distinct attention states, such as focused, distracted, and neutral. The SVM model is trained and optimized using a cross-validation approach to achieve optimal performance and generalization ability.

The performance of the proposed system is evaluated using a comprehensive set of metrics, including accuracy, precision, recall, and F1-score, to assess the classifier's ability to discern different attention levels accurately. The results demonstrate the efficacy of the SVM-based approach in accurately identifying child attention states based on facial expressions. The developed system offers valuable insights into understanding children's attentional patterns and can be utilized in various practical applications, such as educational settings, child psychology, and child-care centers. Additionally, it lays the groundwork for further research and development of more sophisticated models to assess attention-related behaviors in children, potentially leading to early detection and intervention of attention-related disorders. However, challenges related to the generalization of the model across diverse cultural and demographic contexts need to be addressed to enhance its applicability and inclusiveness.

Article Details