Recognition and estimation of relative position of objects in controlled environments

This project presents a system for recognition and classification of objects that are in its environment, for an assistance robot, as well as the estimation of their relative position with respect to the robot. For the recognition and classification of objects, we apply artificial vision techniques...

Descripción completa

Detalles Bibliográficos
Autores: Luna Taylor, Jorge Enrique, Clemente Rosas, Eloy Antonio, Gómez Torres , José Luis, Villa Medina, Isaac
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2022
País:México
Institución:UNIVERSIDAD AUTÓNOMA DEL ESTADO DE HIDALGO
Repositorio:PÄDI Boletín Científico de Ciencias Básicas e Ingeniería del ICBI
Idioma:español
OAI Identifier:oai:repository.uaeh.edu.mx:article/9262
Acceso en línea:https://repository.uaeh.edu.mx/revistas/index.php/icbi/article/view/9262
Access Level:acceso abierto
Palabra clave:Computer vision
Convolutional neural networks
Semantic segmentation
Stereoscopic vision
Artificial intelligence
Visión Artificial
Redes neuronales convolucionales
Segmentación semántica
Visión estereoscópica
Inteligencia artificial
Descripción
Sumario:This project presents a system for recognition and classification of objects that are in its environment, for an assistance robot, as well as the estimation of their relative position with respect to the robot. For the recognition and classification of objects, we apply artificial vision techniques based on semantic segmentation tools, such as convolutional neural networks. For the estimation of the relative position of the objects, once identified, a stereoscopic vision technique was implemented. The results of the experiments show a 90.6% accuracy in recognition and classification, and an average error of 5 cm when estimating the relative position of the objects.