Facial Expression Recognition Based on GSO Enhanced Deep Learning in IOT Environment

Facial expressions play an important role in human communication and integrating deep learning techniques into Internet of Things (IoT) scenarios enhances the understanding of this data, enabling applications in industries such as healthcare, security, and human-computer interaction.The existing methods suffer from lower accuracy and higher computational complexity compared to the proposed Deep Convolutional Neural Network using Galactic Swarm Optimization (DCNN-GSO) approach hinder their practical applicability in real-time image processing tasks.This paper proposes a comprehensive framework for facial expression analysis in IoT environments.The preprocessing phase uses a Gaussian filter to improve image quality and reduce noise.Feature extraction is performed using a spatial temporal interest point (STIP), which captures the spatial and temporal cue of facial expressions.The proposed method leverages Deep Convolutional Neural Networks (DCNN) to extract discriminative features from facial images captured by IoT devices.Galactic Swarm Optimization (GSO) is employed to optimize the hyperparameters of the DCNN model, thereby improving its performance in facial expression classification tasks.By integrating GSO with deep learning, the proposed approach aims to overcome the challenges of limited computational resources and energy constraints inherent in IoT environments.GSO optimizes the parameters of deep learning models for facial expression recognition, improving accuracy and robustness.The proposed framework provides a comprehensive approach for facial expression analysis in IoT environments, solving challenges such as noise, computational complexity, accuracy, etc. in IoT systems, and opening the way for humans and improved device performance and connectivity.The DCNN-GSO method outperforms the competition with remarkable results: 94% accuracy, 92.3% precision, and 91% recall.With a very low mean absolute error (MAE) of 3.46, it shows itself to be a reliable and accurate solution for practical uses.