Diseño e implementación de un guante traductor de señas para personas con discapacidad auditiva y/o verbal.
Gesture-based communication is crucial for people with hearing or speech disabilities. However, at the present time systems are often expensive or difficult to use. This project is focused on the development of a mobile application for gesture interpretation using a dense neural network. The system...
Saved in:
| Main Author: | |
|---|---|
| Format: | bachelorThesis |
| Language: | spa |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://dspace.unl.edu.ec/jspui/handle/123456789/31512 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Add Tag | Summary: | Gesture-based communication is crucial for people with hearing or speech disabilities. However, at the present time systems are often expensive or difficult to use. This project is focused on the development of a mobile application for gesture interpretation using a dense neural network. The system includes a circuit designed to capture the user's hand signals using flexible sensors and an MPU6050, sending the perceived information to a database in Firebase to be processed in an application programmed in Flutter. The mobile application was developed with Dart language, which has allowed the incorporation of main functionalities: visualization of gestures in letters, pronunciation of the interpreted characters and, teaching of gestures through images and videos on platforms such as YouTube. The results indicate a sixty six percent accuracy in the interpretation of global gestures of the trained model, which can be improved by modifying certain training parameters and expanding the database of each class of gestures. Key words: Translator glove; Assisted Technologies; Microcontroller; Sign language; Machine learning; Flutter |
|---|