Toddler Tracking System with Face Recognition and Object Tracking Using Deep Neural Network

Abstract

Toddlers tend to have approach objects that may be dangerous to them at home because of their natural curiosity. When families are sleeping or doing housework, unpredictable injuries may be happening. Horrible, irreversible accidents may occur in less than no time. Not only for this reason but also with the technological evolution, systems related digital parenting have gained importance. Smart technology is adopted by modern parents to provide their children safety. Considering all these situations, the need to design a toddler tracking system for helping the parents has emerged. In this article, a toddler tracking system using Deep Neural Network has been proposed. The proposed system is based on face recognition and object tracking algorithms and created using pre-trained neural networks. The system is based on recognizing the toddlers' faces and following all toddler's movements in the house. When the toddler is getting close dangerous places, tools, furniture, etc, the system alerts the user with the warning system. The proposed system is tested by using the toddlers' data which is collected before and 80.7% test accuracy has been obtained. The experimental result showed that the proposed method has achieved sufficient performance to be compared with state-of-the-art studies.