|
|
|
|
| LEADER |
04240nam a22002057a 4500 |
| 003 |
HR-ZaFER |
| 005 |
20220315145149.0 |
| 008 |
220315b ||||| |||| 00| 0 eng d |
| 999 |
|
|
|c 52694
|d 52694
|
| 040 |
|
|
|a HR-ZaFER
|b hrv
|c HR-ZaFER
|e ppiak
|
| 041 |
|
|
|a eng
|
| 100 |
|
|
|9 43041
|a Šeketa, Goran
|
| 245 |
|
|
|a Evaluation of algorithms for human fall detection based on acceleration measurements :
|b doctoral thesis /
|c mentor Ratko Magjarević
|
| 260 |
|
|
|a Zagreb :
|b G. Šeketa ; Faculty of electrical engineering and computing,
|c 2022.
|
| 300 |
|
|
|a 120 str. :
|b ilustr. u bojama ;
|c 30 cm +
|e CD-ROM
|
| 504 |
|
|
|a Bibliografija : 95 - 111.
|
| 520 |
|
|
|a Falls, in particular falls among elderly, are a major public health problem. Automatic fall detection systems ensure that elderly people receive prompt assistance after experiencing a fall and can reduce the negative consequences the fall. The thesis focused on resolving problems related to the evaluation of fall detection algorithms on simulated datasets by conducting research on three specific issues. First, experiments were conducted to explore the influence of three variabilities in the experiment environment on the acquired acceleration data. The results revealed that all three variabilities cause heterogeneity between data in different fall detection datasets and based on these findings, a protocol for acquiring simulated falls with accelerometers was recommended. Next, a fall detection algorithm was designed and implemented in order to explore the influence of different window configurations in event-centered data segmentation on the performance of the detection algorithm. Based on the results, a method for implementing the data segmentation stage was proposed. Finally, a fall detection algorithm was trained with different numbers of subjects and trials to explore the influence on the performance outcomes. It was found that with more amount of data available for training, the performance follows an exponential raise and saturated at some point. Based on the saturation points, a recommendation for evaluating the relevance of existing fall detection datasets was made.
The original scientific contribution of the thesis is: a dataset acquisition procedure for unbiased evaluation of fall detection algorithms, a method for selecting thresholds in fall detection algorithms and evaluation of relevance of the dataset for the efficacy of fall detection algorithms.
|
| 520 |
|
|
|a Padovi, posebice padovi starijih osoba, predstavljaju značajan javnozdravstveni problem. Sustavi za automatsku detekciju pada omogućuju da se osobi nakon pada pruži brza pomoć i smanje negativne posljedice. Tema doktorskog rada usmjerena je na rješavanje problema povezanih s vrednovanjem algoritama za detekciju pada na skupovima podataka sa simuliranim padovima provedbom istraživanja u tri dijela. Prvo su izvedeni eksperimenti kako bi se istražio utjecaj tri varijabilnosti u načinu prikupljanja skupova podataka na karakteristike mjerenih signala akceleracije. Pronađena je povezanost sve tri istraživane varijabilnosti s mjerenim signalima te je na temelju rezultata predložena metoda za prikupljanje skupova podataka. U drugom dijelu istraživanja projektiran je i implementiran algoritam za detekciju padova. Ispitan je utjecaj različitih konfiguracija prozora za segmentaciju ulaznih podataka na učinkovitost algoritma. Temeljem rezultata, određene su vrijednosti pragova kojima se ostvaruju konfiguracije prozora s najvećom učinkovitosti. Konačno, algoritam za detekciju padova treniran je s različitim brojem uzoraka (brojem ispitanika i brojem ponavljanja pokreta). Otkrivena je eksponencijalna ovisnost učinkovitosti algoritama o broju uzoraka. Prema točkama zasićenja u eksponencijalnoj funkciji, predložen je način vrednovanja skupova podataka sa simuliranim padovima.
Ostvareni znanstveni doprinos je: postupak prikupljanja skupova podataka za nepristrano vrednovanje algoritama za detekciju pada zasnovanih na mjerenju akceleracije,
metoda odabira pragova u algoritmima za detekciju pada te ocjena reprezentativnosti skupa podataka za uspješnost algoritama za detekciju pada.
|
| 700 |
|
|
|4 ths
|9 5701
|a Magjarević, Ratko
|
| 942 |
|
|
|2 udc
|c D
|