Back to search

Πανεπιστήμιο Μακεδονίας

Αναγνώριση δακτυλικών χειρονομιών σε πραγματικό χρόνο για την αλληλεπίδραση ανθρώπου-ρομπότ: εφαρμογή στον έλεγχο συστημάτων μηχατρονικής

Abstract

dc:description

This thesis presents a methodology for real time interaction with mechatronic systems via gesture recognition. It presents a new kind of Human - Machine interaction through personal computer. With the help of real time gesture recognition technologies and by using camera signal processing (web, kinect, pmd) the interaction with systems robotics and mechatronics in general is achieved. The methodology is based on the definition of the appropriate control commands (dictionary of movements) through gestures. It also supports their real-time detection through the orbits of fingertips, even before completion, to achieve interaction with the mechatronic systems or robotics. Part of the methodology is the simultaneous control of multiple commands via finger gestures meaning that each finger represents a different command performed simultaneously.So in the broader context of Human Computer Interaction, research for this thesis aims to present a) an appropriate methodology to create a dictionary of movements b) methodology of command recovery with machine vision technologies in real -time and c) suitable software, FingerOrasis, for the interaction with mechatronic system in real time.The methodology used is based on expanding in three-dimensional space the current pc vision method to recognize finger movements in music. It takes into account two approaches of recognition in real –time from the FingerOrasis system: a) the static approach or touch finger recovery and b) the dynamic approach, command recovery. Static recognition is applied to each video frame. The dynamic recognition applies to all video frame sequences, based on Hidden Markov Models and dynamic time warping technique.

Degree

thesis:*
Grantor dc:publisher
Πανεπιστήμιο Μακεδονίας
Year dc:date
2014

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Tsagkaris, Apostolos
  • Τσαγκάρης, Απόστολος

Subjects

dc:subject × 14

Rights

dc:rights
Statement dc:rights
  • BY_NC_ND
Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/34919

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Tsagkaris, Apostolos; Τσαγκάρης, Απόστολος. Αναγνώριση δακτυλικών χειρονομιών σε πραγματικό χρόνο για την αλληλεπίδραση ανθρώπου-ρομπότ: εφαρμογή στον έλεγχο συστημάτων μηχατρονικής. Πανεπιστήμιο Μακεδονίας, 2014. http://hdl.handle.net/10442/hedi/34919