Vehicle detection in urban environments scanned by a lidar

Alfonso Ramírez-Pedraza, José Joel González-Barbosa, Francisco Javier Ornelas-Rodríguez, Angel Iván García-Moreno, Adan Salazar-Garibay, Erick Alejandro González-Barbosa

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

© 2014 CEA. Publicado por Elsevier España, S.L.U. Todos los derechos reservados. Detection of vehicles on 3D point clouds is performed by using the algorithm presented in this work. Point clouds correspond to urban environments and were acquired with the LIDAR Velodyne HDL-64E. The environment is considered semi-structured so that can be modeled using planes. Vehicle detection is carried out on to stages, segmentation and indexation. First stage is at the same time composed of three sub-stages. In the first one the principal plane (in this case the floor) is extracted, in the second sub-stage secondary planes are extracted using a tailored version of Hough's method, secondary planes are those perpendicular to the main plane. Finally in the third sub-stage and using MeanShift method, the remaining objects are segmented. Indexation on its side is divided into two sub-stages, in the first one, last segmented objects using MeanShift method are modeled using histograms according to the direction of the object's 3D points normal; in the second stage histograms are compared to those previously stored on a database of object's histograms. Optimizing of detection thresholds was carried out through ROC analysis. Two databases were used during the experiments, the first DB have 4500 objects and was used for ROC analysis training; the second one contained 3000 objects and was used for verification.
Original languageAmerican English
Pages (from-to)189-198
Number of pages10
JournalRIAI - Revista Iberoamericana de Automatica e Informatica Industrial
DOIs
StatePublished - 1 Jan 2015
Externally publishedYes

Fingerprint

Optical radar
Experiments

Cite this

Ramírez-Pedraza, Alfonso ; González-Barbosa, José Joel ; Ornelas-Rodríguez, Francisco Javier ; García-Moreno, Angel Iván ; Salazar-Garibay, Adan ; González-Barbosa, Erick Alejandro. / Vehicle detection in urban environments scanned by a lidar. In: RIAI - Revista Iberoamericana de Automatica e Informatica Industrial. 2015 ; pp. 189-198.
@article{f5fb9cc6c433425eba6341d1c7e5f93c,
title = "Vehicle detection in urban environments scanned by a lidar",
abstract = "{\circledC} 2014 CEA. Publicado por Elsevier Espa{\~n}a, S.L.U. Todos los derechos reservados. Detection of vehicles on 3D point clouds is performed by using the algorithm presented in this work. Point clouds correspond to urban environments and were acquired with the LIDAR Velodyne HDL-64E. The environment is considered semi-structured so that can be modeled using planes. Vehicle detection is carried out on to stages, segmentation and indexation. First stage is at the same time composed of three sub-stages. In the first one the principal plane (in this case the floor) is extracted, in the second sub-stage secondary planes are extracted using a tailored version of Hough's method, secondary planes are those perpendicular to the main plane. Finally in the third sub-stage and using MeanShift method, the remaining objects are segmented. Indexation on its side is divided into two sub-stages, in the first one, last segmented objects using MeanShift method are modeled using histograms according to the direction of the object's 3D points normal; in the second stage histograms are compared to those previously stored on a database of object's histograms. Optimizing of detection thresholds was carried out through ROC analysis. Two databases were used during the experiments, the first DB have 4500 objects and was used for ROC analysis training; the second one contained 3000 objects and was used for verification.",
author = "Alfonso Ram{\'i}rez-Pedraza and Gonz{\'a}lez-Barbosa, {Jos{\'e} Joel} and Ornelas-Rodr{\'i}guez, {Francisco Javier} and Garc{\'i}a-Moreno, {Angel Iv{\'a}n} and Adan Salazar-Garibay and Gonz{\'a}lez-Barbosa, {Erick Alejandro}",
year = "2015",
month = "1",
day = "1",
doi = "10.1016/j.riai.2015.03.003",
language = "American English",
pages = "189--198",
journal = "RIAI - Revista Iberoamericana de Automatica e Informatica Industrial",
issn = "1697-7912",
publisher = "Elsevier Doyma",

}

Vehicle detection in urban environments scanned by a lidar. / Ramírez-Pedraza, Alfonso; González-Barbosa, José Joel; Ornelas-Rodríguez, Francisco Javier; García-Moreno, Angel Iván; Salazar-Garibay, Adan; González-Barbosa, Erick Alejandro.

In: RIAI - Revista Iberoamericana de Automatica e Informatica Industrial, 01.01.2015, p. 189-198.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Vehicle detection in urban environments scanned by a lidar

AU - Ramírez-Pedraza, Alfonso

AU - González-Barbosa, José Joel

AU - Ornelas-Rodríguez, Francisco Javier

AU - García-Moreno, Angel Iván

AU - Salazar-Garibay, Adan

AU - González-Barbosa, Erick Alejandro

PY - 2015/1/1

Y1 - 2015/1/1

N2 - © 2014 CEA. Publicado por Elsevier España, S.L.U. Todos los derechos reservados. Detection of vehicles on 3D point clouds is performed by using the algorithm presented in this work. Point clouds correspond to urban environments and were acquired with the LIDAR Velodyne HDL-64E. The environment is considered semi-structured so that can be modeled using planes. Vehicle detection is carried out on to stages, segmentation and indexation. First stage is at the same time composed of three sub-stages. In the first one the principal plane (in this case the floor) is extracted, in the second sub-stage secondary planes are extracted using a tailored version of Hough's method, secondary planes are those perpendicular to the main plane. Finally in the third sub-stage and using MeanShift method, the remaining objects are segmented. Indexation on its side is divided into two sub-stages, in the first one, last segmented objects using MeanShift method are modeled using histograms according to the direction of the object's 3D points normal; in the second stage histograms are compared to those previously stored on a database of object's histograms. Optimizing of detection thresholds was carried out through ROC analysis. Two databases were used during the experiments, the first DB have 4500 objects and was used for ROC analysis training; the second one contained 3000 objects and was used for verification.

AB - © 2014 CEA. Publicado por Elsevier España, S.L.U. Todos los derechos reservados. Detection of vehicles on 3D point clouds is performed by using the algorithm presented in this work. Point clouds correspond to urban environments and were acquired with the LIDAR Velodyne HDL-64E. The environment is considered semi-structured so that can be modeled using planes. Vehicle detection is carried out on to stages, segmentation and indexation. First stage is at the same time composed of three sub-stages. In the first one the principal plane (in this case the floor) is extracted, in the second sub-stage secondary planes are extracted using a tailored version of Hough's method, secondary planes are those perpendicular to the main plane. Finally in the third sub-stage and using MeanShift method, the remaining objects are segmented. Indexation on its side is divided into two sub-stages, in the first one, last segmented objects using MeanShift method are modeled using histograms according to the direction of the object's 3D points normal; in the second stage histograms are compared to those previously stored on a database of object's histograms. Optimizing of detection thresholds was carried out through ROC analysis. Two databases were used during the experiments, the first DB have 4500 objects and was used for ROC analysis training; the second one contained 3000 objects and was used for verification.

U2 - 10.1016/j.riai.2015.03.003

DO - 10.1016/j.riai.2015.03.003

M3 - Article

SP - 189

EP - 198

JO - RIAI - Revista Iberoamericana de Automatica e Informatica Industrial

JF - RIAI - Revista Iberoamericana de Automatica e Informatica Industrial

SN - 1697-7912

ER -