Detekcija pulsa na temelju videosnimke osobe

master's thesis

• Title (english): Detection of Pulse Based on a Video Recording of a Person
• Author: Roko Srđan Buča
• Mentor: Marko Subašić (mentor)
• Committee member: Marko Subašić (predsjednik povjerenstva)
• Committee member: Zvonko Kostanjčar (član povjerenstva)
• Committee member: Tomislav Petković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-02-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom diplomskom radu se obrađuje problem detekcije pulsa potpuno beskontaktnim metodama – naročito analizom videosnimke.

Analiziraju se različiti pristupi problemu, njihove prednosti i mane te potencijalni načini za daljnja poboljšanja.

Obrađuju se algoritmi iz domena bliskih centralnom problemu s posebnim naglaskom na primjenjivost postojećih i budućih algoritama za poboljšanje procesa mjerenja pulsa iz videosnimke osobe, poput načina pojačavanja videosnimke, detekcije lica te rukovanja šumom.

Predstavlja se Python implementacija jednog od mogućih pristupa problemu. Algoritam na temelju promjene intenziteta boje kože kroz vrijeme raznim matematičkim postupcima konačno mjeri šum.

Diskutiraju se rezultati predstavljene implementacije te moguća poboljšanja i poželjni budući radovi na algoritmu.

Abstract (english)

Central topic of this thesis is the analysis of the heartbeat detection methods using only contactless data collectors, namely cameras.

Analysis of existing different approaches to the problem is provided, with the discussion of positive and negative sides of every approach. Potential future improvements are also discussed.

There is an additional analysis of domains relevant to the central thesis with emphasis on provided added usability of the main pulse detection algorithm. Notable areas are face detection, video amplification and noise handling.

Implementation of one such algorithm is provided in Python. Algorithm works by tracking changes in color intensity of the human skin over time. Various mathematical methods are applied to the measured signal in order to calculate the pulse rate.

Implementation's results are provided and discussed, as well as possible future improvements to the algorithm.

• Keywords: Web kamera; Python; detekcija lica; obrada signala; filtriranje šuma; stvarno vrijeme; krvožilni sustav; vremenska analiza
• Keywords (english): Web camera; Python; face detection; signal analysis; noise filtering; real-time; cardiovascular system; time-domain analysis
• Language: croatian
• URN:NBN: urn:nbn:hr:168:879943
• 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: Closed access
• Created on: 2024-10-12 20:12:42