Examinando por Materia "Object detection"

Mostrando 1 - 3 de 3
  • No hay miniatura disponible
    Ítem
    Detección de infracciones y matrículas en motocicletas, mediante visión artificial, aplicado a sistemas inteligentes de transporte
    (Associacao Iberica de Sistemas e Tecnologias de Informacao, 2020-06-01) VALENCIA, JESÚS FERNANDO; Ramirez-Guerrero, T.; Castañeda, L.; Toro, M.; Universidad EAFIT. Departamento de Ingeniería Mecánica; Estudios en Mantenimiento (GEMI)
    Incomplete operation of transportation systems has led to an increase in illegal passenger transportation, by the means of motorcycles, at worldwide, causing high accident rates. In the intermediate cities of Colombia, the main control method applied to informal transportation is to impose fines to identified offenders by transit authorities. This work proposes the development of an application that, by means of computer vision, serves as a tool for transit officers in the detections of three infringements types committed by motorcyclists: not wearing a helmet; circulate in prohibited areas; transporting passenger in sites where it’s not allowed. Our case study was Valledupar city, where 105 motorcyclists’ images were taken while they are driving. Results show an accuracy of 87,5% in infringements detection, showing the relevance of this application as an assistance tool to disincentive informal transportation. © 2020, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.
  • No hay miniatura disponible
    Ítem
    Detección de infracciones y matrículas en motocicletas, mediante visión artificial, aplicado a sistemas inteligentes de transporte
    (Associacao Iberica de Sistemas e Tecnologias de Informacao, 2020-06-01) VALENCIA, JESÚS FERNANDO; Ramirez-Guerrero, T.; Castañeda, L.; Toro, M.; VALENCIA, JESÚS FERNANDO; Ramirez-Guerrero, T.; Castañeda, L.; Toro, M.; Universidad EAFIT. Departamento de Ingeniería de Sistemas; I+D+I en Tecnologías de la Información y las Comunicaciones
    Incomplete operation of transportation systems has led to an increase in illegal passenger transportation, by the means of motorcycles, at worldwide, causing high accident rates. In the intermediate cities of Colombia, the main control method applied to informal transportation is to impose fines to identified offenders by transit authorities. This work proposes the development of an application that, by means of computer vision, serves as a tool for transit officers in the detections of three infringements types committed by motorcyclists: not wearing a helmet; circulate in prohibited areas; transporting passenger in sites where it’s not allowed. Our case study was Valledupar city, where 105 motorcyclists’ images were taken while they are driving. Results show an accuracy of 87,5% in infringements detection, showing the relevance of this application as an assistance tool to disincentive informal transportation. © 2020, Associacao Iberica de Sistemas e Tecnologias de Informacao. All rights reserved.
  • Miniatura
    Ítem
    Evaluation of Robust Covariance Estimation for Object Detection
    (Universidad EAFIT, 2021-04-07) Tamayo-Arango, Andres Felipe; Plazas Escudero, David; Vidal-Correa, Juan Pablo; Universidad EAFIT, School of Sciences, Department of Mathematical Sciences
    This work presents an initial approach to the evaluation of robust covariance estimation for object detection (localization) using the “region covariance” technique from the literature. The covariance estimation is performed using the Comedian, Kendall, Spearman and Ledoit and Wolf robust approaches for covariance, and the procedure was also compared using two different matrix norms for estimating dissimilarity. The performance was measured quantitatively using linear regression and Pareto boundaries, yielding the Ledoit and Wolf estimation with best overall performance in object detection in normal and noisy images.