Metoda klasifikacije i prepoznavanja položaja pri padu i nekim pokretima iz dnevnog života

undergraduate thesis

• Title (english): The method of classification and recognition of position during a fall and some movements from everyday life
• Author: Vida Krušić
• Mentor: Ratko Magjarević (mentor)
• Committee member: Ratko Magjarević (predsjednik povjerenstva)
• Committee member: Davorin Ambruš (član povjerenstva)
• Committee member: Krunoslav Jurčić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-06-30, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ideja ovog rada bila je istražiti mogućnost poboljšanja klasifikacije i točnosti prepoznavanja položaja tijela pri simuliranom padu i sjedanju odnosno ustajanju. Istraženo je jeli moguće to poboljšanje ostvariti uporabom waveleta. Prvo je bilo potrebno odabrati vrstu waveleta koja će se koristiti te je za to korišten programski jezik Matlab u kojemu je nekoliko testnih signala AVM-a transformirano različitim wavelet funkcijama te su uspoređivani kako bi se vidjelo koja najbolje ističe trenutak pada. To je Rickerov, tzv. „Mexican hat“ wavelet koji je u programskom jeziku Python primijenjen na sve ulazne signale te je pomoću njega na svakom signalu pronađen, odnosno na nekima nije pronađen, pad. Rezultati su pokazali da je ovakva metoda određivanja trenutka pada precizna u više od 97% slučajeva što je dovoljno da bi bilo korisno za uporabu u stvarnim sustavima koji bi se koristili za detekciju pada što bi moglo biti korisno pri skrbi o starijim ili ozlijeđenim osobama koje nemaju mogućnost cjelodnevne skrbi.

Abstract (english)

The idea of this work was to investigate the possibility of improving the classification and recognition accuracy of the body position during a simulated fall, sitting and standing up. It was investigated whether it is possible to achieve this improvement by using wavelets. First, it was necessary to choose the type of wavelet to be used. For this the Matlab programming language was used, in which several test signals of the AVM were transformed by different wavelet functions and compared to see which best highlights the moment of fall. It is Ricker's, the so-called The "Mexican hat" wavelet, which was applied in the Python programming language to all input signals. Using it, the fall was found, or missed, on each signal. The results showed that this method of determining the moment of a fall is accurate in more than 97% of cases, which is enough to be useful for use in real systems that would be used for fall detection, which could be useful in the care of elderly or injured people who do not have the possibility of all-day care.

• Keywords: wavelet; AVM; prepoznavanje pokreta tijela
• Keywords (english): wavelet; AVM; body movement recognition
• Language: croatian
• URN:NBN: urn:nbn:hr:168:307252
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-07-13 10:14:19