國立虎尾科技大學 |

Vision-based object tracking using 3-D scanners with applications in educational laboratories.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Vision-based object tracking using 3-D scanners with applications in educational laboratories./
Author:	Zhang, Mingshao.
Description:	1 online resource (182 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 78-07(E), Section: B.
Subject:	Mechanical engineering. -
Online resource:	click for full text (PQDT)
ISBN:	9781369571981

Vision-based object tracking using 3-D scanners with applications in educational laboratories.
Zhang, Mingshao.

Vision-based object tracking using 3-D scanners with applications in educational laboratories. - 1 online resource (182 pages)

Source: Dissertation Abstracts International, Volume: 78-07(E), Section: B.

Thesis (Ph.D.)--Stevens Institute of Technology, 2016.

Includes bibliographical references

Motion capture is the process of recording the movement of objects or people, which is often used in military, entertainment, sports and medical applications as well as for the validation of computer vision algorithms and robotics systems. The focus of this dissertation is on the tracking of the motion of rigid bodies using 3-D scanners. This process includes detecting, recognizing, localizing and tracking objects in a scene. It is believed that these techniques have great potential in applications such as vehicle navigation, robot control, human-computer interaction, etc.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9781369571981Subjects--Topical Terms:

557493
Mechanical engineering.
Index Terms--Genre/Form:

554714
Electronic books.

Vision-based object tracking using 3-D scanners with applications in educational laboratories.
LDR:04192ntm a2200361K 4500 001 915259
005 20180727125211.5
006 m o u
007 cr mn||||a|a||
008 190606s2016 xx obm 000 0 eng d
020 $a 9781369571981
035 $a (MiAaPQ)AAI10190660
035 $a (MiAaPQ)stevens:10348
035 $a AAI10190660
040 $a MiAaPQ $b eng $c MiAaPQ
100 1 $a Zhang, Mingshao. $3 1188564
245 1 0 $a Vision-based object tracking using 3-D scanners with applications in educational laboratories.
264 0 $c 2016
300 $a 1 online resource (182 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 78-07(E), Section: B.
500 $a Adviser: Sven K. Esche.
502 $a Thesis (Ph.D.)--Stevens Institute of Technology, 2016.
504 $a Includes bibliographical references
520 $a Motion capture is the process of recording the movement of objects or people, which is often used in military, entertainment, sports and medical applications as well as for the validation of computer vision algorithms and robotics systems. The focus of this dissertation is on the tracking of the motion of rigid bodies using 3-D scanners. This process includes detecting, recognizing, localizing and tracking objects in a scene. It is believed that these techniques have great potential in applications such as vehicle navigation, robot control, human-computer interaction, etc.
520 $a Instead of reconstructing traditional hardware setups for motion tracking, the research work presented here focuses more on software approaches (i.e. algorithm development and modifications). In the first part of this document, existing vision-based object tracking techniques will be introduced and traditional computer vision algorithms related to motion tracking will be studied and tested. Depending on the application scenarios, the software system will be constructed by modifying traditional algorithms (if they are suitable) or developing new ones (if no suitable candidates exist). The ultimate goal is to develop a general, reliable and robust software system that can be used to track various objects with the help of different 3-D scanners.
520 $a A three stage software system will be constructed (by modifying traditional computer vision algorithms and integrating originally piece-wise algorithms into a functioning system), implemented and tested using a low-cost 3-D scanner (i.e. the Microsoft Kinect). The performance and remaining shortcomings of this system will be studied. In order to overcome these remaining shortcomings and further improve the system's performance, a thorough study of the technical specifics of the scanner (the Kinect) will be introduced first. An observation of the inconsistency between the assumption of traditional computer vision algorithms and the scanners' unique characteristics will be introduced and its potential influence on the tracking results will be discussed. In order to address the observed problem, a systematic algorithm modification approach (i.e. modification of traditional algorithms according to the unique properties of the 3-D scanner) will be developed, implemented and tested.
520 $a The software system and modified algorithms are able to produce acceptable results. However, the relatively complex setup procedure and required extensive preparation represent a limitation for potential applications. In order to mediate these concerns, a new technique for vision-based object tracking was developed, which involves algorithms that can simultaneously track the motion and reconstruct the model of objects. In order to validate the tracking system, the algorithm modification technique and the new tracking method, an innovative game-based desktop Virtual Reality educational laboratory was developed and implemented.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Mechanical engineering. $3 557493
650 4 $a Robotics. $3 561941
655 7 $a Electronic books. $2 local $3 554714
690 $a 0548
690 $a 0771
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Stevens Institute of Technology. $b Mechanical Engineer. $3 1178936
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10190660 $z click for full text (PQDT)