Embedded system application development on Raspberry Pi 3 - machine learning household inventory tracking
Keeping track of household inventory can be a time-consuming and tedious task. Having passed by the supermarket and you have no idea which groceries/items at home are running low and required to be topped up. What if you have “eyes” inside your cabinet to tell you the information that you need? This...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/166850 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Keeping track of household inventory can be a time-consuming and tedious task. Having passed by the supermarket and you have no idea which groceries/items at home are running low and required to be topped up. What if you have “eyes” inside your cabinet to tell you the information that you need? This is very much practical and convenient in our daily lives thanks to modern technologies. Millions of people's lives could be improved by this by easing the stress of keeping track of household goods and freeing up time for more pleasurable pursuits.
The objective of this project is to develop an object-detecting inventory tracking system with the Raspberry Pi 3 model. The system uses a camera to capture images of household items and applies YOLOv4 tiny object detection to identify the objects in the images. Having identified the objects, the system will be able to quantify the objects in each category. The inventory tracking will be updated periodically to the Microsoft Azure IoT hub, and users will be able to access the data remotely when they need it. |
|---|