Human Activity Recognition and Fall Detection

This project aims to detect human activities and identify falls in real time using machine learning and sensor data. It contributes to elderly care and health monitoring by enabling prompt alerts and intervention when a fall occurs.

“The system can be integrated into healthcare applications, smart homes, or IoT platforms for proactive fall prevention and health tracking.”

• 

  ▶

The system uses data from wearable sensors (e.g., accelerometers and gyroscopes) to classify physical activities such as walking, sitting, standing, running, and falling. Various machine learning and deep learning models were evaluated, with the selected model achieving high accuracy in real-world testing. The solution can be embedded in smartwatches or mobile devices for continuous monitoring and alerts, enhancing safety and independence for elderly or vulnerable individuals.