Novelty Detection and Segmentation based on Gaussian Mixture Models: a Case Study in 3D Robotic Laser Mapping

P. Drews Jr., P. Núñez, R. P. Rocha, M. Campos and J. Dias

Robotics and Autonomous Systems, 61(12), pp. 1696-1709, Elsevier, Dec. 2013. DOI dx.doi.org/10.1016/j.robot.2013.06.004

Abstract

This article proposes a framework to detect and segment changes in robotics datasets, using 3D robotic mapping as a case study. The problem is very relevant in several application domains, not necessarily related with mobile robotics, including security, health, industry and military applications. The aim is to identify significant changes by comparing current data with previous data provided by sensors. This feature is extremely challenging because large amounts of noisy data must be processed in a feasible way. The proposed framework deals with novelty detection and segmentation in robotic maps using clusters provided by Gaussian Mixture Models (GMMs). GMMs provide a feature space that enables data compression and effective processing. Two alternative criteria to detect changes in the GMM space are compared: a greedy technique based on the Earth Mover’s Distance (EMD); and a structural matching algorithm that fulfills both absolute (global matching) and relative constraints (structural matching). The proposed framework is evaluated with real robotic datasets and compared with other methods known from literature. With this purpose, 3D mapping experiments are carried out with both simulated data and real data from a mobile robot equipped with a 3D range sensor.

Keywords ─ Novelty detection, Gaussian mixture model, 3D robotic mapping.

Full text

You may ask Rui P. Rocha for an electronic copy of this publication’s full text by e-mail:
.

Please select for your message’s subject ‘Requesting Rui Rocha’s electronic copy’ and include on the message’s body your full name, title and affiliation, why do you need to access the publication and the BibTeX information below.

BibTeX

@ARTICLE(Drews_et_al_13,

AUTHOR = "Drews Jr., P. and N{\’u}{\~n}ez, P. and Rocha, R. P. and Campos, M. and Dias, J.",

TITLE = "Novelty Detection and Segmentation based on Gaussian Mixture Models: a Case Study in 3D Robotic Laser Mapping",

JOURNAL = "Robotics and Autonomous Systems",

VOLUME = "61",

NUMBER = "12",

YEAR = "2013",

MONTH = "Dec.",

PAGES = "1696-1709"

)

(top of the page)

Last update: 30/10/2013