Identifikacija oseb na osnovi gibanja telesa, zajetega z napravo Kinect

diplomsko delo

Peter Zlodej (Author), Boris Cigale (Mentor), Sebastijan Šprager (Co-mentor)

Abstract

V diplomskem delu je predstavljen postopek identifikacije oseb na podlagi posnetkov človeške hoje, zajetih z napravo Kinect. Opisan je proces zajemanja signalov, predobdelave podatkov in strojnega učenja z metodo podpornih vektorjev. Predstavljeni so tudi rezultati testiranj uspešnosti postopka. Ugotovili smo, da je na podlagi signalov hoje, zajetih z napravo Kinect, možno razpoznati osebo iz množice petih oseb z natančnostjo 93,13 ± 9,99 %.

Keywords

identifikacija oseb;senzor Kinect;analiza hoje;metoda podpornih vektorjev;biometrija;

Data

Language:	Slovenian
Year of publishing:	2013
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[P. Zlodej]
UDC:	004.8:004.932(043.2)
COBISS:	17526806
Views:	1359
Downloads:	150
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	IDENTIFICATION OF PERSONS USING FULL-BODY MOTION DATA CAPTURED WITH KINECT
Secondary abstract:	This diploma presents a person identification method based on human gait signals, captured with Kinect sensor. It describes signal capture process, data preprocessing and machine learning with support vector machine. Test results of the proposed method are also presented in the document. Results show, that it is possible to recognize a person from a group of five persons, with accuracy of 93.13 ± 9.99 %, using human gait signals, captured with Kinect.
Secondary keywords:	person identification;Kinect sensor;gait analysis;support vector machine;biometrics;
URN:	URN:SI:UM:
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko
Pages:	VIII, 28 f.
ID:	8727100