The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Les robots de service doivent être capables de reconnaître et d’identifier des objets situés dans des arrière-plans complexes. Puisqu’aucune méthode ne peut fonctionner à elle seule dans toutes les situations, plusieurs méthodes doivent être combinées et les robots doivent sélectionner automatiquement celle qui convient. Dans cet article, nous proposons un schéma pour classer les situations en fonction des caractéristiques de l'objet d'intérêt et de la demande des utilisateurs. Nous classons les situations en quatre groupes et employons des techniques différentes pour chacun. Nous utilisons la transformation de caractéristiques invariantes d'échelle (SIFT), l'analyse des composantes principales du noyau (KPCA) en conjonction avec la machine à vecteurs de support (SVM) en utilisant les fonctionnalités d'intensité, de couleur et de Gabor pour cinq catégories d'objets. Nous montrons que l’utilisation de fonctionnalités appropriées est importante pour l’utilisation de techniques basées sur KPCA et SVM sur différents types d’objets. Grâce à des expériences, nous montrons qu'en utilisant notre schéma de catégorisation, un robot de service peut sélectionner une fonctionnalité et une méthode appropriées, et améliorer considérablement ses performances de reconnaissance. Pourtant, la reconnaissance n’est pas parfaite. Ainsi, nous proposons de combiner la méthode autonome avec une méthode interactive qui permet au robot de reconnaître la demande de l'utilisateur pour un objet et une classe spécifiques lorsque le robot ne parvient pas à reconnaître l'objet. Nous proposons également une manière interactive de mettre à jour le modèle objet utilisé pour reconnaître un objet en cas de panne, en conjonction avec les commentaires de l'utilisateur.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Al MANSUR, Yoshinori KUNO, "Specific and Class Object Recognition for Service Robots through Autonomous and Interactive Methods" in IEICE TRANSACTIONS on Information,
vol. E91-D, no. 6, pp. 1793-1803, June 2008, doi: 10.1093/ietisy/e91-d.6.1793.
Abstract: Service robots need to be able to recognize and identify objects located within complex backgrounds. Since no single method may work in every situation, several methods need to be combined and robots have to select the appropriate one automatically. In this paper we propose a scheme to classify situations depending on the characteristics of the object of interest and user demand. We classify situations into four groups and employ different techniques for each. We use Scale-invariant feature transform (SIFT), Kernel Principal Components Analysis (KPCA) in conjunction with Support Vector Machine (SVM) using intensity, color, and Gabor features for five object categories. We show that the use of appropriate features is important for the use of KPCA and SVM based techniques on different kinds of objects. Through experiments we show that by using our categorization scheme a service robot can select an appropriate feature and method, and considerably improve its recognition performance. Yet, recognition is not perfect. Thus, we propose to combine the autonomous method with an interactive method that allows the robot to recognize the user request for a specific object and class when the robot fails to recognize the object. We also propose an interactive way to update the object model that is used to recognize an object upon failure in conjunction with the user's feedback.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e91-d.6.1793/_p
Copier
@ARTICLE{e91-d_6_1793,
author={Al MANSUR, Yoshinori KUNO, },
journal={IEICE TRANSACTIONS on Information},
title={Specific and Class Object Recognition for Service Robots through Autonomous and Interactive Methods},
year={2008},
volume={E91-D},
number={6},
pages={1793-1803},
abstract={Service robots need to be able to recognize and identify objects located within complex backgrounds. Since no single method may work in every situation, several methods need to be combined and robots have to select the appropriate one automatically. In this paper we propose a scheme to classify situations depending on the characteristics of the object of interest and user demand. We classify situations into four groups and employ different techniques for each. We use Scale-invariant feature transform (SIFT), Kernel Principal Components Analysis (KPCA) in conjunction with Support Vector Machine (SVM) using intensity, color, and Gabor features for five object categories. We show that the use of appropriate features is important for the use of KPCA and SVM based techniques on different kinds of objects. Through experiments we show that by using our categorization scheme a service robot can select an appropriate feature and method, and considerably improve its recognition performance. Yet, recognition is not perfect. Thus, we propose to combine the autonomous method with an interactive method that allows the robot to recognize the user request for a specific object and class when the robot fails to recognize the object. We also propose an interactive way to update the object model that is used to recognize an object upon failure in conjunction with the user's feedback.},
keywords={},
doi={10.1093/ietisy/e91-d.6.1793},
ISSN={1745-1361},
month={June},}
Copier
TY - JOUR
TI - Specific and Class Object Recognition for Service Robots through Autonomous and Interactive Methods
T2 - IEICE TRANSACTIONS on Information
SP - 1793
EP - 1803
AU - Al MANSUR
AU - Yoshinori KUNO
PY - 2008
DO - 10.1093/ietisy/e91-d.6.1793
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E91-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2008
AB - Service robots need to be able to recognize and identify objects located within complex backgrounds. Since no single method may work in every situation, several methods need to be combined and robots have to select the appropriate one automatically. In this paper we propose a scheme to classify situations depending on the characteristics of the object of interest and user demand. We classify situations into four groups and employ different techniques for each. We use Scale-invariant feature transform (SIFT), Kernel Principal Components Analysis (KPCA) in conjunction with Support Vector Machine (SVM) using intensity, color, and Gabor features for five object categories. We show that the use of appropriate features is important for the use of KPCA and SVM based techniques on different kinds of objects. Through experiments we show that by using our categorization scheme a service robot can select an appropriate feature and method, and considerably improve its recognition performance. Yet, recognition is not perfect. Thus, we propose to combine the autonomous method with an interactive method that allows the robot to recognize the user request for a specific object and class when the robot fails to recognize the object. We also propose an interactive way to update the object model that is used to recognize an object upon failure in conjunction with the user's feedback.
ER -