Multiple object tracking for robust quantitative analysis of passenger motion while boarding and alighting a metropolitan train

José Sebastián Gómez Meza, José Delpiano, Sergio A. Velastin, Rodrigo Fernández, Sebastián Seriani Awad

Resultado de la investigación: Contribución a una conferenciaArtículorevisión exhaustiva

Resumen

To achieve significant improvements in public transport it is necessary to develop an autonomous system that locates and counts passengers in real time in scenarios with a high level of occlusion, providing tools to efficiently solve problems such as reduction and stabilization in travel times, greater fluency, better control of fleets and less congestion. A deep learning method based in transfer learning is used to accomplish this: You Only Look Once (YOLO) version 3 and Faster RCNN Inception version 2 architectures are fine tuned using PAMELA-UANDES dataset, which contains annotated images of the boarding and alighting of passengers on a subway platform from a superior perspective. The locations given by the detector are passed through a multiple object tracking system implemented based on a Markov decision process that associates subjects in consecutive frames and assigns identities considering overlaps between past detections and predicted positions using a Kalman filter.

Idioma originalInglés
Páginas231-238
Número de páginas8
DOI
EstadoPublicada - 7 oct 2021
Evento11th International Conference of Pattern Recognition Systems, ICPRS 2021 - Virtual, Online
Duración: 17 mar 202119 mar 2021

Conferencia

Conferencia11th International Conference of Pattern Recognition Systems, ICPRS 2021
CiudadVirtual, Online
Período17/03/2119/03/21

Nota bibliográfica

Publisher Copyright:
© 2021 Institution of Engineering and Technology. All rights reserved.

Huella

Profundice en los temas de investigación de 'Multiple object tracking for robust quantitative analysis of passenger motion while boarding and alighting a metropolitan train'. En conjunto forman una huella única.

Citar esto