| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 17.8 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
This dissertation presents the development of a prototype pill dispenser with the aim of combating the common problem of non-compliance and lack of adherence to medication regimes. In order to make the device more autonomous, the growing capabilities of neural networks and the computing power of microcomputers such as the Raspberry Pi were explored.
The integration of Facial Recognition and Speech Recognition models into these microcomputers was investigated. These solutions offer significant improvements, including greater security and a more accessible interaction experience for users.
During the course of this project, Facial Recognition models were trained, and their performance evaluated using robust metrics such as the F1-Score and the ROC curve. We also studied how these models can be effectively applied to a drug dispensing system and microcomputers. In addition, a comprehensive analysis of the application of pre-trained Speech Recognition models was carried out.
This work represents an important step towards improving the effectiveness and accessibility of drug dispensing systems, contributing to greater patient adherence to their medication regimes.
Description
Keywords
Inteligência Artificial Deep Learning Facial Recognition Raspberry Pi Speech Recognition