Abstract

In our constantly aging society, that already today shows a lack of nursing personnel, it becomes of more and more importance to offer elderly citizens the possibility of an autonomous life. Therefore this thesis deals with the effective usage of modern technologies like machine learning and augmented reality in applications for elderly people. For this, two IOS applications are developed. The first of them classifies different pills though a trained YOLOv5 object detection network and marks them in an augmented reality scene. Additionally it offers the possibility to livestream the mobile applications screen to a nurse or caregiver, who can point out the right pill remotely. The second app offers the possibility to scan EAN-13 barcodes on groceries and gives information about how healthy those are. In an preset profile, users can define nutritional restrictions, allergies or diets which are considered in the recommendation. Both approaches are designed with a most usable user interface, and a survey to this environment is conducted.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]