DATA STREAMING AND VISUALIZATION THE RESULTS OF ARTIFICIAL INTELLIGENCE SYSTEM TO DETECT PLASTIC BOTTLE WASTE
Plastic bottle waste is one of the biggest contributors to plastic waste in the world. Plastic waste in Indonesian rivers and oceans also dominated by plastic bottle waste, even though plastic bottle has a high recycling value. The large amount of waste in Indonesian rivers, makes plastic bott...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/55468 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Plastic bottle waste is one of the biggest contributors to plastic waste in the world.
Plastic waste in Indonesian rivers and oceans also dominated by plastic bottle
waste, even though plastic bottle has a high recycling value. The large amount of
waste in Indonesian rivers, makes plastic bottle waste hard to detect and count. The
rivers flow also makes the counting results hard to follow.
The goals of this final project are to create a system that can detect and count the
plastic bottle waste in real-time. This system is also expected to visualize the
detection and counting results to a platform also in real-time. This system is created
using artificial intelligence with You Only Look Once the third version object
detection algorithm. The platform that used to visualize the detection results is a
website with a front-end created using HTML and CSS, back-end created using
JavaScript, CanvasJS, and Ajax, and Flask is used as the web server.
The test results show that the system accuration is affected by the bottle condition,
the bottle position, and the other objects placed with the bottles. The visualization
system can already visualize the detection and counting results as a number and a
graph. |
|---|