Desarrollo de un sistema para el control de asistencia estudiantil mediante el uso de algoritmos de reconocimiento facial basado en visión artificial
Facial recognition has attracted great interest in various industries due to its multiple applications and benefits. In education, its implementation significantly simplifies attendance taking compared to traditional methods. This work focuses on developing a face recognition-based attendance system...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | bachelorThesis |
| 語言: | spa |
| 出版: |
2024
|
| 主題: | |
| 在線閱讀: | https://dspace.unl.edu.ec/jspui/handle/123456789/30319 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
添加標簽 | 總結: | Facial recognition has attracted great interest in various industries due to its multiple applications and benefits. In education, its implementation significantly simplifies attendance taking compared to traditional methods. This work focuses on developing a face recognition-based attendance system that adapts to the specific needs of teachers and students. The project uses pre-trained Deep Learning models such as VGGFace, Facenet and ArcFace, with Facenet showing the best results. To recognize students with a minimum number of images per person, Siamese Networks were used, which made it possible to identify new users without having to retrain the model. The creation and deployment of the system involved the use of databases, user managers and the development of a desktop application with the Python QtDesigner library, allowing its use by both teachers and students. The results of the system in a real environment show an overall accuracy of 87.1%, highlighting the effectiveness of the selected model despite the inherent challenges. This paper not only addresses the practical application of face recognition in educational environments, but also highlights the importance of considering ethical and legal aspects to ensure the correct implementation and acceptance of the technology. Keywords: Face recognition, attendance control, computer vision, Python |
|---|