Inspection of four-sensor falls detector

Journal Title: Bulletin of the Military University of Technology - Year 2015, Vol 64, Issue 2

Abstract

The studies presented in this article are the continuation of previous work to develop a mobile fall detector. The algorithm is based on a discrete wavelet transform of the signals from the sensors available at the detector and a linear support vector machine as a classifier. Fisher score method is used for feature selection in the proposed algorithm. As a result of reducing the number of features, the number of support vectors has been also reduced — it has a direct impact on the upper estimate of the classification error. On the basis of the obtained results, the classifier parameters have been calculated. This allows presenting the developed concept in the field of ROCROCROC curves (Receiver Operating Characteristics) and their comparison with the results obtained for individual sensors. The developed concept gives much better results than each of the sensors acting independently. The findings of this study have given very good results in comparison with the previous findings, with a significant reduction in the number of required features. Due to the close relationship between the number of training data and the number of support vectors which directly affect the upper estimate of the classification error, the number of features has been reduced. Finally, satisfactory results have been obtained with the reduction of the number of features from 38 to just six, ensuring that the upper estimation of the classification error in the set of the new test data does not exceed 5.3%.[b]Keywords[/b]: falls detection, data fusion, discrete wavelet transform, support vector machine

Authors and Affiliations

Bartłomiej Wójtowicz, Andrzej P. Dobrowolski

Keywords

Related Articles

Badania smarności olejów zawierających heksagonalny azotek boru

W pracy przedstawiono wyniki badań właściwości smarnościowych oleju hydraulicznego Hydrol L-HL 32, oleju przekładniowego Mobilube 1 SHC 75w-90 oraz dwóch olejów bazowych SN-150 i SN-650, zawierających heksagonalny azotek...

Wpływ parametrów CBR gruntów wątpliwych na wymiarowanie nawierzchni lotniskowych

W artykule przedstawiono zagadnienie dotyczące doboru grubości warstw dla projektowanych betonowych nawierzchni lotniskowych w zależności od wartości wyznaczonych parametrów obliczeniowych podłoża gruntowego. Analiza ta...

Diagnosing of a complex technical object in four-valued logic

This paper presents the essence of an investigation of a complex technical object with the use of four-valued logic. To this end, an intelligent diagnostic system (DIAG 2) is described. A special feature of this system w...

Minkowski metrics in creating universal ranking algorithms

The paper presents a general procedure for creating the rankings of a set of objects, while the relation of preference based on any ranking function. The analysis was possible to use the ranking functions began by showin...

Wzmacnianie nawierzchni lotniskowych warstwami z mieszanek mineralno-asfaltowych

W artykule przedstawiono oryginalne rozwiązania oraz metodykę postępowania przy wyznaczaniu grubości nakładek z mieszanek mineralno-asfaltowych dla remontów oraz modernizacji istniejących nawierzchni lotniskowych.Wykorzy...

Download PDF file
  • EP ID EP62765
  • DOI -
  • Views 45
  • Downloads 0

How To Cite

Bartłomiej Wójtowicz, Andrzej P. Dobrowolski (2015). Inspection of four-sensor falls detector. Bulletin of the Military University of Technology, 64(2), 45-58. https://europub.co.uk/articles/-A-62765