Inspection process for dimensioning through images and fuzzy logic

This paper presents a hybrid methodology based on a type 1 fuzzy model in singleton version using a 2k factorial design that optimizes the model of the expert system and serves to perform in-line inspection. The factorial design method provides the required database for the creation of the rule base for the fuzzy model and also generates the database to train the expert system. The proposed method was validated in the process of verifying dimensional parameters by means of images compared with the ANFIS and RBFN models which show greater margins of error in the approximation of the function represented by the system compared with the proposed model. The results obtained show that the model has an excellent performance in the prediction and quality control of the industrial process studied when compared with similar expert system techniques as ANFIS and RBFN.

• Autores: amelec, viloria; Pineda Lezama, Omar Bonerge; Cabrera, Danelys
• Tipo de recurso: Article of journal
• Fecha de publicación: 2020
• Institución: Corporación Universidad de la Costa
• Repositorio: REDICUC - Repositorio CUC
• Idioma: eng
• OAI Identifier: oai:repositorio.cuc.edu.co:11323/7664
• Acceso en línea: https://hdl.handle.net/11323/7664; https://doi.org/10.1016/j.procs.2020.07.095; https://repositorio.cuc.edu.co/
• Palabra clave: Neural networks; Irrigation control; Instrumentation and image analysis; Micro-greenhouse
• Rights: openAccess
• License: CC0 1.0 Universal