Sistema de detección y clasificación de piezas defectuosas mediante visión por computador e inteligencia artificial
Fecha
2023-10-25
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
La detección temprana de defectos en la fabricación de piezas y su posterior clasificación en función de la
gravedad de los mismos constituye un factor elemental en la correcta implementación de políticas de control de
calidad en la industria.
En este trabajo, se propone el diseño y desarrollo de un sistema automatizado que, por medio de técnicas de
visión por computador, permita identificar defectos e imperfecciones en la fabricación de piezas inyectadas en
plástico, dentro de un sistema industrial. Posteriormente, el sistema permitirá clasificar dichos defectos en
función de su gravedad, usando para ello algoritmos de segmentación clásicos de visión por computador, junto
con técnicas de Inteligencia Artificial.
The early detection of defects in the manufacturing of pieces and their subsequent classification based on their severity constitutes an elementary factor in the correct implementation of quality control policies in the industry. In this work, the design and development of an automated system is proposed that, through computer vision techniques, allows identifying defects and imperfections in the manufacturing of plastic injected pieces, within an industrial system. Subsequently, the system will allow these defects to be classified based on their severity, using classic computer vision segmentation algorithms, along with Artificial Intelligence techniques.
The early detection of defects in the manufacturing of pieces and their subsequent classification based on their severity constitutes an elementary factor in the correct implementation of quality control policies in the industry. In this work, the design and development of an automated system is proposed that, through computer vision techniques, allows identifying defects and imperfections in the manufacturing of plastic injected pieces, within an industrial system. Subsequently, the system will allow these defects to be classified based on their severity, using classic computer vision segmentation algorithms, along with Artificial Intelligence techniques.