Bibliographic citations
Nuñez, V., Melgarejo, A. (2024). Desarrollo de una máquina clasificadora de objetos huecos para la automatización del proceso de control de calidad de las pecanas en las plantaciones de Ica [Trabajo de suficiencia profesional, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/674837
Nuñez, V., Melgarejo, A. Desarrollo de una máquina clasificadora de objetos huecos para la automatización del proceso de control de calidad de las pecanas en las plantaciones de Ica [Trabajo de suficiencia profesional]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2024. http://hdl.handle.net/10757/674837
@misc{renati/419595,
title = "Desarrollo de una máquina clasificadora de objetos huecos para la automatización del proceso de control de calidad de las pecanas en las plantaciones de Ica",
author = "Melgarejo Reyes, Arnold Piero",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2024"
}
The development of a sorting machine for Mahan Pecans is presented, which are produced in the south of Lima, specifically in the department of Ica. For the development of the sorting machine, the sound detection method will be used. A stainless-steel plate will receive the impact of the pecan, and the sound produced by the impact will be collected by a high-sensitivity microphone. To achieve better accuracy in detecting the pecan's core, the pecan will pass through a camera before the sound detection method, where its area will be determined through image processing. Then, a microcontroller will identify the characteristics of the sound produced by the pecans and include the area as an additional feature. After completing the mentioned two phases, a Support Vector Machine (SVM) classifier will be used to classify the pecans through an algorithm. Following the detection process, the machine will have a separator that will be activated according to the final classification result. In this project, 100 pecan samples will be used to train the SVM, which will be evenly divided between those without kernels and those with kernels. To evaluate the system's efficiency, 100 randomly selected pecans from both types will be used. The obtained accuracy percentage is 95%, classifying 20 pecans per minute.
This item is licensed under a Creative Commons License