IMPLEMENTASI AUGMENTED REALITY PADA WINDOWS PHONE
Augmented Reality (AR), a technology that combines virtual object and real environment in 3D, is hardly ran smoothly in mobile devices because it is computationally expensive to make AR run real-time. More powerful devices is needed. Smartphone that has been coming in recent years has faster process...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Theses and Dissertations NonPeerReviewed |
| Published: |
[Yogyakarta] : Universitas Gadjah Mada
2014
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/128975/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=69350 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universitas Gadjah Mada |
| Summary: | Augmented Reality (AR), a technology that combines virtual object and real environment in 3D, is hardly ran smoothly in mobile devices because it is computationally expensive to make AR run real-time. More powerful devices is needed. Smartphone that has been coming in recent years has faster processor and bigger memory than previous mobile devices is expected to run AR smoothly in real-time.
In this study, an Augmented Reality System will be built on Windows Phone which is one of operating system that runs in smartphone. The system will be built based on processes used in ARToolkitPlus, an Augmented Reality library. In the implementation a library called AForge will be used on some process such as quadrilateral detection and estimation pose. Then, some test are performed to verify the correctness of marker detection and pose generated by system also time needed to detect markers will be measured.
Test results show that the smallest image size, 160x120, gives faster marker detection than the other two image size which is 640x480 and 320x240, but the correctness of marker detection is decreasing as the number of markers grow. Using dynamic threshold in marker pattern check process can improve the detection resulting markers that are previously not detected in the smallest image size are now correctly detected. |
|---|