Home > Research > Publications & Outputs > Automatic scene recognition for low-resource de...
View graph of relations

Automatic scene recognition for low-resource devices using evolving classifiers

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paper

Published

Standard

Automatic scene recognition for low-resource devices using evolving classifiers. / Andreu, Javier; Dutta Baruah, Rashmi; Angelov, Plamen.

2011 IEEE International Conference on Fuzzy Systems (FUZZ). IEEE, 2011. p. 2779-2785.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paper

Harvard

Andreu, J, Dutta Baruah, R & Angelov, P 2011, Automatic scene recognition for low-resource devices using evolving classifiers. in 2011 IEEE International Conference on Fuzzy Systems (FUZZ). IEEE, pp. 2779-2785, IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011, Taipei, Taiwan, Province of China, 19/11/11. https://doi.org/10.1109/FUZZY.2011.6007720

APA

Andreu, J., Dutta Baruah, R., & Angelov, P. (2011). Automatic scene recognition for low-resource devices using evolving classifiers. In 2011 IEEE International Conference on Fuzzy Systems (FUZZ) (pp. 2779-2785). IEEE. https://doi.org/10.1109/FUZZY.2011.6007720

Vancouver

Andreu J, Dutta Baruah R, Angelov P. Automatic scene recognition for low-resource devices using evolving classifiers. In 2011 IEEE International Conference on Fuzzy Systems (FUZZ). IEEE. 2011. p. 2779-2785 https://doi.org/10.1109/FUZZY.2011.6007720

Author

Andreu, Javier ; Dutta Baruah, Rashmi ; Angelov, Plamen. / Automatic scene recognition for low-resource devices using evolving classifiers. 2011 IEEE International Conference on Fuzzy Systems (FUZZ). IEEE, 2011. pp. 2779-2785

Bibtex

@inproceedings{7dd196c058444b92a2456ced2beae04c,
title = "Automatic scene recognition for low-resource devices using evolving classifiers",
abstract = "In this paper an original approach is proposed which makes possible autonomous scenes recognition performed on-line by an evolving self-learning classifier. Existing approaches for scene recognition are off-line and used in intelligent albums for picture categorization/selection. The emergence of powerful mobile platforms with camera on board and sensor-based autonomous (robotic) systems is pushing forward the requirement for efficient self-learning and adaptive/evolving algorithms. Fast real-time and online algorithms for categorisation of the real world environment based on live video stream are essential for understanding and situation awareness as well as for localization and context awareness. In scene analysis the critical problem is feature extraction mechanism for a quick description of the scene. In this paper we apply a well known technique called spatial envelop or GIST. Visual scenes can be quite different but very often they can be grouped in similar types/categories. For example, pictures from different cities across the Globe, e.g. Tokyo, Vancouver, New York Moscow, Dusseldorf, etc. bear the similar pattern of an urban scene high rise buildings, despite the differences in the architectural style. Same applies for the beaches of Miami, Maldives, Varna, Costa del Sol, etc. One assumption based on which such automatic video classifiers can be build is to pre-train them using a large number of such images from different groups. Variety of possible scenes suggests the limitations of such an approach. Therefore, we use in this paper the recently propose evolving fuzzy rule-based classifier, simpleClass, which is self learning and thus updates its rules and categories descriptions with each new image. In addition, it is fully recursive, computationally efficient and yet linguistically transparent.",
keywords = "Human activity recognition , Fuzzy classifiers, Evolving systems, Wearable sensors, accelerometers",
author = "Javier Andreu and {Dutta Baruah}, Rashmi and Plamen Angelov",
year = "2011",
month = sep,
doi = "10.1109/FUZZY.2011.6007720",
language = "English",
isbn = "978-1-4244-7315-1 ",
pages = "2779--2785",
booktitle = "2011 IEEE International Conference on Fuzzy Systems (FUZZ)",
publisher = "IEEE",
note = "IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011 ; Conference date: 19-11-2011",

}

RIS

TY - GEN

T1 - Automatic scene recognition for low-resource devices using evolving classifiers

AU - Andreu, Javier

AU - Dutta Baruah, Rashmi

AU - Angelov, Plamen

PY - 2011/9

Y1 - 2011/9

N2 - In this paper an original approach is proposed which makes possible autonomous scenes recognition performed on-line by an evolving self-learning classifier. Existing approaches for scene recognition are off-line and used in intelligent albums for picture categorization/selection. The emergence of powerful mobile platforms with camera on board and sensor-based autonomous (robotic) systems is pushing forward the requirement for efficient self-learning and adaptive/evolving algorithms. Fast real-time and online algorithms for categorisation of the real world environment based on live video stream are essential for understanding and situation awareness as well as for localization and context awareness. In scene analysis the critical problem is feature extraction mechanism for a quick description of the scene. In this paper we apply a well known technique called spatial envelop or GIST. Visual scenes can be quite different but very often they can be grouped in similar types/categories. For example, pictures from different cities across the Globe, e.g. Tokyo, Vancouver, New York Moscow, Dusseldorf, etc. bear the similar pattern of an urban scene high rise buildings, despite the differences in the architectural style. Same applies for the beaches of Miami, Maldives, Varna, Costa del Sol, etc. One assumption based on which such automatic video classifiers can be build is to pre-train them using a large number of such images from different groups. Variety of possible scenes suggests the limitations of such an approach. Therefore, we use in this paper the recently propose evolving fuzzy rule-based classifier, simpleClass, which is self learning and thus updates its rules and categories descriptions with each new image. In addition, it is fully recursive, computationally efficient and yet linguistically transparent.

AB - In this paper an original approach is proposed which makes possible autonomous scenes recognition performed on-line by an evolving self-learning classifier. Existing approaches for scene recognition are off-line and used in intelligent albums for picture categorization/selection. The emergence of powerful mobile platforms with camera on board and sensor-based autonomous (robotic) systems is pushing forward the requirement for efficient self-learning and adaptive/evolving algorithms. Fast real-time and online algorithms for categorisation of the real world environment based on live video stream are essential for understanding and situation awareness as well as for localization and context awareness. In scene analysis the critical problem is feature extraction mechanism for a quick description of the scene. In this paper we apply a well known technique called spatial envelop or GIST. Visual scenes can be quite different but very often they can be grouped in similar types/categories. For example, pictures from different cities across the Globe, e.g. Tokyo, Vancouver, New York Moscow, Dusseldorf, etc. bear the similar pattern of an urban scene high rise buildings, despite the differences in the architectural style. Same applies for the beaches of Miami, Maldives, Varna, Costa del Sol, etc. One assumption based on which such automatic video classifiers can be build is to pre-train them using a large number of such images from different groups. Variety of possible scenes suggests the limitations of such an approach. Therefore, we use in this paper the recently propose evolving fuzzy rule-based classifier, simpleClass, which is self learning and thus updates its rules and categories descriptions with each new image. In addition, it is fully recursive, computationally efficient and yet linguistically transparent.

KW - Human activity recognition

KW - Fuzzy classifiers

KW - Evolving systems

KW - Wearable sensors

KW - accelerometers

U2 - 10.1109/FUZZY.2011.6007720

DO - 10.1109/FUZZY.2011.6007720

M3 - Conference contribution/Paper

SN - 978-1-4244-7315-1

SP - 2779

EP - 2785

BT - 2011 IEEE International Conference on Fuzzy Systems (FUZZ)

PB - IEEE

T2 - IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011

Y2 - 19 November 2011

ER -