DocumentCode :
575539
Title :
Scene classification using unsupervised neural networks for mobile robot vision
Author :
Madokoro, Hirokazu ; Utsumi, Yuya ; Sato, Kazuhito
Author_Institution :
Dept. of Machine Intell. & Syst. Eng., Akita Prefectural Univ., Akita, Japan
fYear :
2012
fDate :
20-23 Aug. 2012
Firstpage :
1568
Lastpage :
1573
Abstract :
This paper presents an unsupervised scene classification method based on context of features for semantic recognition of indoor scenes used for an autonomous mobile robot. Our method creates Visual Words (VWs) of two types using Scale-Invariant Feature Transform (SIFT) and Gist. Using the combination of VWs, our method creates Bags of VWs (BoVWs) to vote to a two-dimensional histogram as context-based features. Moreover, our method generates labels as a candidate of categories with maintaining stability and plasticity together using the incremental learning function of Adaptive Resonance Theory-2 (ART-2). Our method actualizes unsupervised learning based scene classification using generated labels of ART-2 for teaching signals of Counter Propagation Networks (CPNs). The spatial and topological relations among scenes are mapped on the category map of CPNs. The relations of classified scenes that contain categories are visualized on the category map. The experiment demonstrates that classification accuracy of semantic categories such as office rooms, corridors, etc. using an open dataset for an evaluation platform of position estimation and navigation for an autonomous mobile robot.
Keywords :
image classification; mobile robots; neural nets; path planning; robot vision; unsupervised learning; 2D histogram; ART-2; adaptive resonance theory 2; autonomous mobile robot vision; context based feature; counter propagation networks; gist; incremental learning function; indoor scenes; navigation; plasticity; position estimation; scale invariant feature transform; semantic category; semantic recognition; unsupervised learning; unsupervised neural networks; unsupervised scene classification; visual words; Accuracy; Context; Feature extraction; Histograms; Mobile robots; Semantics; ART; Gist; Robot Vision; SIFT; SOM;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
SICE Annual Conference (SICE), 2012 Proceedings of
Conference_Location :
Akita
ISSN :
pending
Print_ISBN :
978-1-4673-2259-1
Type :
conf
Filename :
6318701
Link To Document :
بازگشت