Bibliographic citations
This is an automatically generated citacion. Modify it if you see fit
Del, E., (2018). Desarrollo de un sistema de visión artificial para realizar una clasificación uniforme de limones [Tesis, Universidad Privada del Norte]. https://hdl.handle.net/11537/13118
Del, E., Desarrollo de un sistema de visión artificial para realizar una clasificación uniforme de limones [Tesis]. PE: Universidad Privada del Norte; 2018. https://hdl.handle.net/11537/13118
@misc{sunedu/2808128,
title = "Desarrollo de un sistema de visión artificial para realizar una clasificación uniforme de limones",
author = "Del Castillo Huaccha, Eduardo",
publisher = "Universidad Privada del Norte",
year = "2018"
}
Title: Desarrollo de un sistema de visión artificial para realizar una clasificación uniforme de limones
Authors(s): Del Castillo Huaccha, Eduardo
Advisor(s): Salazar Campos, Juan Orlando
OCDE field: https://purl.org/pe-repo/ocde/ford#2.02.04
Issue Date: 13-Mar-2018
Institution: Universidad Privada del Norte
Abstract: La presente tesis planteó como objetivo principal la creación de un sistema de visión
artificial que permita realizar una clasificación uniforme de limones. Las formas y
dimensiones de los limones a ser analizados están sujetos al códex de la lima-limón de la
Organización de Comida y Agricultura de las Naciones Unidas.
Actualmente se analizó el contexto internacional y nacional determinando que no existe
tecnología de información asociada al proceso de clasificación de limones, esto nos brinda
la posibilidad de explorar alternativas basadas en áreas de la computación que ayuden en
este proceso, tal es el caso de la visión artificial.
Se diseñó una solución siguiendo las fases de visión artificial (adquisición, pre
procesamiento, segmentación, descripción y reconocimiento e interpretación), donde se
realizó una comparativa entre los algoritmos dentro de cada fase, para identificar cuáles
son los que mejor se adaptan a la problemática planteada permitiendo obtener un resultado
adecuado.
Luego de haber desarrollado el sistema de visión artificial aplicando los algoritmos
seleccionados, se obtuvo como resultado que el sistema tiene una eficacia de 83.9%, una
sensibilidad de 82.8% y una especificidad del 100%.
Por otro lado, el tiempo de procesamiento para clasificar un limón es de 0.33 segundos.
Gracias a los resultados obtenidos se pudo comprobar la hipótesis en la que se sostiene
que un sistema de visión artificial permite realizar una clasificación uniforme de limones.
This thesis main objective is to create an artificial vision system able to do a uniform classification of lemons. Sizes and dimensions of lemons are tied to the Lime-lemon codex from the United Nations Food and Agriculture Organization. Currently the international and national context was analyzed, determining that there is no information technology associated with the lemons classification process, which gives us the possibility to explore alternatives based on areas of computation that helps in this process, such as the artificial vision. A solution was designed applying the algorithms belonging to the phases of the artificial vision (acquisition, pre-processing, segmentation, description and recognition and interpretation). Also, a comparison was made between the algorithms within each phase to identify which are the ones that has a better fit to this system and give an suitable result. After having implemented the artificial vision system, using the selected algorithms, it was obtained as a result that the artificial vision system has an efficiency of 83.9%, a sensitivity of 82.8% and a specificity of 100% On the other hand, the processing time to classify a lemon is of 0.33 seconds. Thanks to the results obtained it was possible to verify the hypothesis in which it is maintained that an artificial vision system is able to perform a uniform classification of lemons.
This thesis main objective is to create an artificial vision system able to do a uniform classification of lemons. Sizes and dimensions of lemons are tied to the Lime-lemon codex from the United Nations Food and Agriculture Organization. Currently the international and national context was analyzed, determining that there is no information technology associated with the lemons classification process, which gives us the possibility to explore alternatives based on areas of computation that helps in this process, such as the artificial vision. A solution was designed applying the algorithms belonging to the phases of the artificial vision (acquisition, pre-processing, segmentation, description and recognition and interpretation). Also, a comparison was made between the algorithms within each phase to identify which are the ones that has a better fit to this system and give an suitable result. After having implemented the artificial vision system, using the selected algorithms, it was obtained as a result that the artificial vision system has an efficiency of 83.9%, a sensitivity of 82.8% and a specificity of 100% On the other hand, the processing time to classify a lemon is of 0.33 seconds. Thanks to the results obtained it was possible to verify the hypothesis in which it is maintained that an artificial vision system is able to perform a uniform classification of lemons.
Link to repository: https://hdl.handle.net/11537/13118
Note: Trujillo San Isidro
Discipline: Ingeniería de Sistemas Computacionales
Grade or title grantor: Universidad Privada del Norte. Facultad de Ingeniería
Grade or title: Ingeniero de Sistemas Computacionales
Juror: Peralta, José Luis; Gutiérrez Magan, Luis; Leiva Via, Geancarlo
Register date: 14-Mar-2018; 14-Mar-2018
This item is licensed under a Creative Commons License