Identification of household electrical appliances - Ghazali, Fateh
40,99 €
inkl. MwSt.
Versandkostenfrei*
Versandfertig in 6-10 Tagen
payback
20 °P sammeln
  • Broschiertes Buch

The aim of this work is to propose a new identification system for domestic electrical appliances. Our first contribution is to propose an identification system based on the use of statistical parameters of harmonics and the application of the KNN classifier combined with the voting rule method. The results obtained show that the extraction of 500 parameters, based on the estimation of the statistical mean and standard deviation, combined with KNN classification and the voting rule strategy, gives the best CR Classification Rate of 94.97%. Our second contribution, is to reduce dimensionality…mehr

Andere Kunden interessierten sich auch für
Produktbeschreibung
The aim of this work is to propose a new identification system for domestic electrical appliances. Our first contribution is to propose an identification system based on the use of statistical parameters of harmonics and the application of the KNN classifier combined with the voting rule method. The results obtained show that the extraction of 500 parameters, based on the estimation of the statistical mean and standard deviation, combined with KNN classification and the voting rule strategy, gives the best CR Classification Rate of 94.97%. Our second contribution, is to reduce dimensionality by using a compact parameter representation (called DWE) that is based on estimating the mean and standard deviation of the energy calculated at each dyadic decomposition level of the wavelet analysis. Two descriptors called LWE and WCC are also extracted from this analysis by applying respectively the logarithm of the Total Energy and the discrete cosine transform. The results show that the WCC descriptor gives a maximum CR of 98.13%.
Autorenporträt
Ghazali Fateh machte 2002 seinen Abschluss als Ingenieur an der Universität Constantine, 2011 seinen Abschluss als Magister an der Universität M'sila und 2022 seinen Doktortitel an der Universität Bordj bou Arréridj. Seine Hauptinteressen gelten der Mustererkennung, Klassifizierung und Identifizierung von elektrischen Geräten.