Design of an automated system for the classification of plastic and glass bottles using sound

Nowadays, The Internet of Things (IoT) is a valuable tool which allows companies to automate and control their processes in a more efficient way by interconnecting hardware and software. For this project, we designed an automated system intended to classify elements in a recycling process. This was done by applying Deep Learning and audio processing techniques, as well as by interconnecting a conveyor band, a robot arm, and an audio frequency sensor via various IoT communication protocols. The result is a fully automated system, which successfully classifies bottles based on the sound made when hit, with an accuracy of over 90%

• Autores: Restrepo Sánchez, Santiago; Sánchez Rodríguez, Juan Camilo
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2022
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/56422
• Acceso en línea: http://hdl.handle.net/1992/56422
• Palabra clave: IoT; MFFCs; Deep Learning; Mel Spectrogram; Audio; recycle; Internet de las cosas; Reciclaje; Aprendizaje automático (Inteligencia artificial); Botellas; Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/