DESIGN AND IMPLEMENTATION OF AUGMENTED REALITY TECHNOLOGY TOWARDS GHOST SPAWNING ON LOCATION BASED APPLICATION: GAME THE DEMITE
Initially, games are played traditionally. However, currently, games can be played using hardware such as smartphones. With today's gaming technology, a lot of types of games have popped up, and one of them is a casual game. The Demite application developed by Bandung Institute of Technol...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/55228 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Initially, games are played traditionally. However, currently, games can be played using
hardware such as smartphones. With today's gaming technology, a lot of types of games have
popped up, and one of them is a casual game. The Demite application developed by Bandung
Institute of Technology students is a video game regarding how to catch ghosts using
pronunciation. This game uses Location-based Augmented Reality Technology where GPS is
needed to create digital maps and find, capture, and fight those ghost objects. This ghost object
also applies augmented reality technology. The Convolutional Neural Network (CNN) method
for placing ghost objects in The Demite application becomes the right solution to determine
where the ghost objects will be placed.
The Test is carried out on digital map, ghost object identification, performance efficiency,
RabbitMQ, GPS accuracy and user satisfaction. The results of digital map test using Unity and
smart phone shows that the tiles which contain data of map information can be implemented
on client and server , map information is obtained from OpenStreetMap. Ghost object
identification test results can be implemented in 3 categories, those are trees, bottles and cars
which have high degree of accuracy. The results of testing and evaluation of the model for 10
experiments on three types of images show the accuracy of trees is 75.79%, bottles 70.95%
and cars 70.83%.Performance efficiency test results show the memory usage ranges from 40
to 80 MB by the CPU usage is about 4.5%. The RabbitMQ test results show that the message
size of RabbitMQ is no more than 40kB. The GPS test results at 50 locations in Bandung show
that the average level of accuracy in the application is 57.72 meters which lack of accuracy.
The results of overall user satisfaction values are 87%. This proves that the game gets positive
response from the users. |
|---|