Algoritam za praćenje, segmentaciju, klasifikaciju i brojanje repetitivnih ljudskih pokreta tijekom tjelovježbe pomoću nosivih senzora

master's thesis

• Title (english): Algorithm for Tracking, Segmentation, Classification and Counting of Repetitive Human Movements During Exercising Using Wearable Sensors
• Author: Marko Čibarić
• Mentor: Ratko Magjarević (mentor)
• Committee member: Ratko Magjarević (predsjednik povjerenstva)
• Committee member: Tomislav Pribanić (član povjerenstva)
• Committee member: Dinko Oletić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Praćenje fizičkih aktivnosti pomoću nosivih senzora ima brojne primjene u medicini i sportu. Specifično područje je i praćenje repetitivnih pokreta. Ono uključuje segmentaciju, klasifikaciju i brojanje napravljenih pokreta što je bio zadatak ovog diplomskog rada. Pokreti su segmentirani i klasificirani iz norme akceleracije. Segmentiranje pokreta je izvršeno pomoću frekvencijskog spektra norme akceleracije dok je klasifikacija rađena pomoću algoritma „dinamičko savijanje vremena“. Za praćenje aktivnosti korišteni su senzorski čvorovi tvrtke Shimmer, a snimljeni podatci su naknadno obrađivani pomoću programskog alata Matlab u kojem je i napisan algoritam. Algoritam je isproban na signalima nekoliko ispitanika.

Abstract (english)

Monitoring physical activities using wearable sensors has numerous applications in medicine and sports. Specific area of monitoring physical activity is monitoring repetitive movements. It includes segmentation, classification and counting of movements that are performed by the examinee, which was the task of this thesis. The movements are segmented and classified from the acceleration norm. Motion segmentation was performed using frequency spectrum of the acceleration norm while the classification was done using the “Dynamic Time Warping” algorithm. Shimmer sensor nodes were used to monitor activity, and the recorded data were subsequently processed using the Matlab software tool, in which the algorithm was designed. The algorithm was tested on signals from several subjects.

• Keywords: fizička aktivnost; repetitivni pokreti; segmentacija; klasifikacija; dinamičko savijanje vremena; nosivi senzori; akceleracija; norma akceleracije; frekvencijski spektar; Matlab; tjelovježba; filtriranje
• Keywords (english): physical activity; repetitive movements; segmentation; classification; dynamic time warping; wearable sensors; acceleration; acceleration norm; frequency spectrum; Matlab; physical exercise; filtering
• Language: croatian
• URN:NBN: urn:nbn:hr:168:503033
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 09:53:27