Deep learning-based garbage bags and potholes detection model using Raspberry Pi

Given the current process for garbage collection and road maintenance, due to gaps in the pavement, in addition to the non-compliance of citizens with the norms established by entities for these services; city streets are becoming dirtier and less passable, affecting transportation. The main problem is that the entities in charge of these tasks do not have daily updated information. In the proposed article, a model for the detection of garbage bags and holes based on artificial vision and deep learning is proposed, which collects geographic information from garbage bags and holes present in the streets of a city. From this information a heat map is generated, which can be provided to the companies in charge of cleaning and maintaining the streets, and contribute to the progress towards a smart city. The behavior of the model has been explored and tested using a Raspberry Pi in real time, and the model has been shown to be fully functional and efficient. The overall performance of the proposed model has been achieved in terms of accuracy, precision and F1-score as 83%, 91% and 82% respectively.

• Autores: Palacios Sánchez, Juan Felipe; Vitery Gómez, Santiago
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2021
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/53500
• Acceso en línea: http://hdl.handle.net/1992/53500
• Palabra clave: Ciudades inteligentes; Redes neuronales (Computadores); Raspberry Pi (Computadora); Aprendizaje automático (Inteligencia artificial); Recolección de basuras; Ingeniería
• Rights: openAccess
• License: https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf