Detección de Turbidez en Plantas de Tratamiento de Agua Potable: Análisis entre YOLOv5x y YOLOv5n
Turbidity is a fundamental parameter for evaluating the quality of drinking water, as it directly affects the effectiveness of treatment processes. This study aims to compare the performance of two YOLOv5 model variants (YOLOv5x and YOLOv5n) in the automated detection of turbidity levels in drinking...
| Autores: | , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | Perú |
| Recursos: | Universidad Nacional Mayor de San Marcos |
| Repositorio: | Revistas - Universidad Nacional Mayor de San Marcos |
| Idioma: | español |
| OAI Identifier: | oai:revistasinvestigacion.unmsm.edu.pe:article/31012 |
| Acesso em linha: | https://revistasinvestigacion.unmsm.edu.pe/index.php/rpcsis/article/view/31012 |
| Access Level: | acceso abierto |
| Palavra-chave: | Object detection Computer vision Turbidity Water quality YOLOv5 Detección de objetos Visión por computadora Turbidez Calidad del agua |
| Resumo: | Turbidity is a fundamental parameter for evaluating the quality of drinking water, as it directly affects the effectiveness of treatment processes. This study aims to compare the performance of two YOLOv5 model variants (YOLOv5x and YOLOv5n) in the automated detection of turbidity levels in drinking water treatment plants (PTAP). A dataset containing representative images of various turbidity levels was used and divided into two subsets: one for training and one for testing, following standard practices in computer vision tasks. The models were evaluated using key metrics such as precision, recall, and F1-Score, the latter being a combined measure that balances precision and sensitivity. YOLOv5x, designed for more complex tasks, demonstrated better overall performance, while YOLOv5n—a lighter variant—showed rapid convergence during training, making it suitable for resource-constrained environments. In conclusion, YOLOv5x is ideal for applications where precision is critical, whereas YOLOv5n is a more efficient option in less demanding scenarios. This study highlights the potential of artificial intelligence to improve water quality monitoring. |
|---|