Fuzzy systems: case study classification of fruit Mc Stipitata Vaug (Áraza)
Abstract
The overall objective of the research was to extract the human expert knowledge and be able to implement it in a programming block that made the human task with the lowest possible error. In this article the objective specific is implement a solution whose base is the logical diffuse with the minor error in the classification of the fruit Mc Stipitata Vaug (Arazá). This research is quantitative, of the applicatif type in the field of computer simulations of experimental processes such as the cultivation of the araza. Is developed a methodology for the determination of the State of maturity of the Araza (Eugenia Stipitata Mc Waugh) based in tools of vision artificial, technical of intelligence computational e implementation in platform FPGA and DSP. Fuzzy logic allows the understanding of different systems in more than two states without changing the working domain, which allows visualizing the system variables and performing different analyzes to make control or classification decisions, which is the case, where A type I fuzzy logic block is used for the classification of the fruit Mc Stiptita Vaug depending on the color of the element, for the implementation of this type of block it is necessary to determine with precision the work domain which the result was used Of the application of filters extracted from statistical analyzes according to classification given by an expert. In The results you can see that the error is low and that the possibility of the emergence of this value may be the causal error originated in the sorting process by a human expert. It is concluded that the error in the sorting process using a random sample inference engine is relatively small, product of the causal error at the time of bullfighting's classification of the fruit, by the human expert.
Downloads
References
Instituto Amazónico de Investigaciones Científicas SINCHI. (2006). Arazá. Bogotá: Universidad Nacional de Colombia.
Luna, G. M. (2002). Introducción a la lógica difusa. México: Centro de Investigación y Estudios Avanzados del IPN.
Mamdani, E. H. (1977). Application of Fuzzy Logic to Approximate Reasoning Using Linguistic Synthesis. IEEE Transactions on Computers, 26(12), 1182 - 1191.
Pastelleto, S. (2012). Diseño de Controladores Fuzzy. Santa Fe - Argentina: Universidad Nacional de Rosario.
Piña, A. B. (2009). Síntesis de Sistemas de Control Borroso Estables por Diseño. Huelva - España: Universidad de Huelva.
Rubio, A. P. (2000). Integración de operadores de implicación y métodos de defuzzificación en sistemas basados en reglas difusas. Implementación, análisis y caracterización. (Tesis Doctoral). Universidad de Granada. Granada - España:
S., P. D., & S., P. F. (2001). La distribución normal. Unidad de Epidemiología Clínica y Bioestadística. Complexo Hospitalario Universitario de A Coruña (8), 268-274.
Vargas, H. F., & Tovar, M. F. (2015). Artificial Vision in Agricultural Products Classification. VII Congreso Internacional de Telemática, 1023 - 1029.
__________ (2016). Desarrollo de una metodología para la determinación del estado de madurez del Arazá (Eugenia Stipitata Mc Vaug) basada en herramientas de visión artificial, técnicas de inteligencia de visión computacional e implementación en plataformas FPGA y DSP. Florencia: Universidad de La Amazonia.