Object Classification and Recognition From Mobile Laser Scanning Point Clouds in a Road Environment

Research output: Contribution to journalArticleScientificpeer-review

Standard

Object Classification and Recognition From Mobile Laser Scanning Point Clouds in a Road Environment. / Lehtomäki, Matti; Jaakkola, Anttoni; Hyyppä, Juha; Lampinen, Jouko; Kaartinen, Harri; Kukko, Antero; Puttonen, Eetu; Hyyppä, Hannu.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 54, No. 2, 02.2016, p. 1226-1239.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Bibtex - Download

@article{7935ce9a6b9b429cb1c35c0ba8234e00,
title = "Object Classification and Recognition From Mobile Laser Scanning Point Clouds in a Road Environment",
abstract = "Automatic methods are needed to efficiently process the large point clouds collected using a mobile laser scanning (MLS) system for surveying applications. Machine-learning-based object recognition from MLS point clouds in a road and street environment was studied in order to create maps from the road environment infrastructure. The developed automatic processing workflow included the following phases: the removal of the ground and buildings, segmentation, segment classification, and object location estimation. Several novel geometry-based features, which were previously applied in autonomous driving and general point cloud processing, were applied for the segment classification of MLS point clouds. The features were divided into three sets, i.e., local descriptor histograms (LDHs), spin images, and general shape and point distribution features, respectively. These were used in the classification of the following roadside objects: trees, lamp posts, traffic signs, cars, pedestrians, and hoardings. The accuracy of the object recognition workflow was evaluated using a data set that contained more than 400 objects. LDHs and spin images were applied for the first time for machine-learning-based object classification in MLS point clouds in the surveying applications of the road and street environment. The use of these features improved the classification accuracy by 9.6{\%} (resulting in 87.9{\%} accuracy) compared with the accuracy obtained using 17 general shape and point distribution features that represent the current state of the art in the field of MLS; therefore, significant improvement in the classification accuracy was achieved. Connected component segmentation and ground extraction were the cause of most of the errors and should be thus improved in the future.",
keywords = "Laser radar, machine vision, remote sensing, URBAN-ENVIRONMENT, EXTRACTION, SEGMENTATION, VEHICLE, NORMALITY, SAMPLES",
author = "Matti Lehtom{\"a}ki and Anttoni Jaakkola and Juha Hyypp{\"a} and Jouko Lampinen and Harri Kaartinen and Antero Kukko and Eetu Puttonen and Hannu Hyypp{\"a}",
note = "VK: Lampinen, J.",
year = "2016",
month = "2",
doi = "10.1109/TGRS.2015.2476502",
language = "English",
volume = "54",
pages = "1226--1239",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
number = "2",

}

RIS - Download

TY - JOUR

T1 - Object Classification and Recognition From Mobile Laser Scanning Point Clouds in a Road Environment

AU - Lehtomäki, Matti

AU - Jaakkola, Anttoni

AU - Hyyppä, Juha

AU - Lampinen, Jouko

AU - Kaartinen, Harri

AU - Kukko, Antero

AU - Puttonen, Eetu

AU - Hyyppä, Hannu

N1 - VK: Lampinen, J.

PY - 2016/2

Y1 - 2016/2

N2 - Automatic methods are needed to efficiently process the large point clouds collected using a mobile laser scanning (MLS) system for surveying applications. Machine-learning-based object recognition from MLS point clouds in a road and street environment was studied in order to create maps from the road environment infrastructure. The developed automatic processing workflow included the following phases: the removal of the ground and buildings, segmentation, segment classification, and object location estimation. Several novel geometry-based features, which were previously applied in autonomous driving and general point cloud processing, were applied for the segment classification of MLS point clouds. The features were divided into three sets, i.e., local descriptor histograms (LDHs), spin images, and general shape and point distribution features, respectively. These were used in the classification of the following roadside objects: trees, lamp posts, traffic signs, cars, pedestrians, and hoardings. The accuracy of the object recognition workflow was evaluated using a data set that contained more than 400 objects. LDHs and spin images were applied for the first time for machine-learning-based object classification in MLS point clouds in the surveying applications of the road and street environment. The use of these features improved the classification accuracy by 9.6% (resulting in 87.9% accuracy) compared with the accuracy obtained using 17 general shape and point distribution features that represent the current state of the art in the field of MLS; therefore, significant improvement in the classification accuracy was achieved. Connected component segmentation and ground extraction were the cause of most of the errors and should be thus improved in the future.

AB - Automatic methods are needed to efficiently process the large point clouds collected using a mobile laser scanning (MLS) system for surveying applications. Machine-learning-based object recognition from MLS point clouds in a road and street environment was studied in order to create maps from the road environment infrastructure. The developed automatic processing workflow included the following phases: the removal of the ground and buildings, segmentation, segment classification, and object location estimation. Several novel geometry-based features, which were previously applied in autonomous driving and general point cloud processing, were applied for the segment classification of MLS point clouds. The features were divided into three sets, i.e., local descriptor histograms (LDHs), spin images, and general shape and point distribution features, respectively. These were used in the classification of the following roadside objects: trees, lamp posts, traffic signs, cars, pedestrians, and hoardings. The accuracy of the object recognition workflow was evaluated using a data set that contained more than 400 objects. LDHs and spin images were applied for the first time for machine-learning-based object classification in MLS point clouds in the surveying applications of the road and street environment. The use of these features improved the classification accuracy by 9.6% (resulting in 87.9% accuracy) compared with the accuracy obtained using 17 general shape and point distribution features that represent the current state of the art in the field of MLS; therefore, significant improvement in the classification accuracy was achieved. Connected component segmentation and ground extraction were the cause of most of the errors and should be thus improved in the future.

KW - Laser radar

KW - machine vision

KW - remote sensing

KW - URBAN-ENVIRONMENT

KW - EXTRACTION

KW - SEGMENTATION

KW - VEHICLE

KW - NORMALITY

KW - SAMPLES

UR - http://dx.doi.org/10.1109/TGRS.2015.2476502

U2 - 10.1109/TGRS.2015.2476502

DO - 10.1109/TGRS.2015.2476502

M3 - Article

VL - 54

SP - 1226

EP - 1239

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 2

ER -

ID: 1730174