Sustav za prepoznavanje aktivnosti osobe u kućanstvu

master's thesis

• Title (english): System for Person Activity Recognition in Household Environment
• Author: Mislav Kovačić
• Mentor: Igor Čavrak (mentor)
• Committee member: Igor Čavrak (član povjerenstva)
• Committee member: Danko Basch (član povjerenstva)
• Committee member: Ivana Bosnić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2025-02-20, Croatia

Abstract

Ovaj rad bavi se razvojem sustava za nadzor aktivnosti starijih osoba koje žive samostalno, uz primjenu neinvazivnih mmWave osjetila. Glavni cilj je detekcija prisutnosti, pokreta i vitalnih znakova (otkucaja srca i ritma disanja), čime se omogućuje brzo uočavanje potencijalno opasnih situacija u kućanstvu. Istražene su tehničke i društvene prepreke uvođenju takvih rješenja, posebno u pogledu privatnosti i prihvaćenosti od strane korisnika. Predstavljeno je središnje računalo za obradu podataka u stvarnom vremenu, frontend sučelje za pregled informacija te simulator aktivnosti za testiranje bez fizičkih osjetila. Sustav nudi modularnost i mogućnost daljnjeg proširenja, primjerice primjenom algoritama strojnog učenja, čime bi se osigurala sveobuhvatnija zaštita i poboljšana kvaliteta praćenja aktivnosti i zdravstvenog stanja korisnika.

Abstract (english)

This paper deals with the development of a system for monitoring the activities of older adults living independently, utilizing non-invasive mmWave sensors. The primary goal is to detect presence, movement, and vital signs (heart rate and breathing rate), thereby enabling the timely detection of potentially hazardous situations in the home environment. Technical and social challenges in introducing such solutions—particularly those related to privacy and user acceptance—have been explored. The paper presents a central computer for real-time data processing, a frontend interface for viewing information, and an activity simulator for testing without the need for physical sensors. The system offers modularity and the possibility of further expansion, for instance through the application of machine learning algorithms, thus providing more comprehensive protection and improved quality in monitoring users’ activities and health status.

• Keywords: pametna kućanstva; starije osobe; mmWave; mmWave osjetila; osjetila; praćenje aktivnosti; detekcija pokreta; vitalni znakovi; nadzor; neinvazivni nadzor; privatnost; Internet stvari; IoT
• Keywords (english): smart homes; elderly; mmWave; mmWave sensors; sensors; activity monitoring; motion detection; vital signs; monitoring; non-invasive monitoring; privacy; Internet of Things; IoT
• Language: croatian
• URN:NBN: urn:nbn:hr:168:383571
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2025-03-12 11:10:32