1Department of Electronics & Communication Engineering, St. Peters Institute of Higher Education and Research, Avadi, Tamil Nadu, India
2Department of Information Science & Engineering Sri Krishna Institute of Technology, Bangalore, India
Corresponding author email: rameshgp@yahoo.com
Article Publishing History
Received: 11/10/2020
Accepted After Revision: 29/12/2020
This project illustrates a method of smart recognition of human behavior to automatically recognize human actions from skeletal joint movements and integrate the skills. This is a low-cost solution and has high precision. An independent mobile app is also intended to track the condition of individuals and their environment while they are alone. The mobile application also incorporates a Notification API integration to enable the sending of warning alerts during irregular conditions. Therefore, our initiative offers a way to assist senior citizens and children with some kind of mishap and health problems. This research proposed novel Convolutional Neural Network (ConvNet/CNN) to predict the action based on human activity. Proposed model provides 3% better accuracy than the existing state of art methodologies.
Abnormal Activity, Action Recognition, Augmented Data, Convolutional Neural Network, Low Cost