Human Activity Recognition Using Convolutional Neural Networks

This research addresses human activity recognition (HAR) using a deep learning framework. Particularly convolutional neural networks (CNNs) to identify and categorize human actions using sensor data. Employing a complete dataset containing numerous actions. It features reclining, sitting, standing, and different walking modes. The study applies CNN models to attain the best precision and accuracy. The models' strong performance in these CNN. Constraints in dataset diversity and size could impact real-world applicability. This article asks for more investigation and design. It combines additional sensors and concentrates emphasis on real-time HAR systems. The paper illustrates the potential of deep learning models in HAR.