A fuzzy RANSAC algorithm based on reinforcement learning concept

Author

Watanabe, Toshio

Author_Institution

Fac. of Eng., Osaka Electro-Commun. Univ., Neyagawa, Japan

fYear

2013

fDate

7-10 July 2013

Firstpage

1

Lastpage

6

Abstract

In the computer vision approach, there are many problems of modeling to prevent affections of noises by sensing units such as cameras and projectors. In order to improve the performance of the modeling in the computer vision, it is necessary to develop a robust modeling technique for various functions. The RANSAC algorithm has been widely applied for such issues. However, the performance is deteriorated when the ratio of noises increases. In this study, a new fuzzy RANSAC algorithm based on the reinforcement learning concept is proposed. The essential performance of the algorithm is evaluated through numerical experiments. From the results, the method is found to be promising to improve calculation time, optimality of the model, and robustness in terms of modeling performance.

Keywords

computer vision; fuzzy set theory; learning (artificial intelligence); random processes; calculation time; computer vision approach; fuzzy RANSAC algorithm; model optimality; modeling performance; noise ratio; numerical experiments; random sample consensus algorithm; reinforcement learning concept; robust modeling technique; sensing units; Algorithm design and analysis; Computational modeling; Computer vision; Data models; Estimation; Learning (artificial intelligence); Noise; RANSAC; computer vision; fuzzy set; reinforcement learning; robust estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems (FUZZ), 2013 IEEE International Conference on

Conference_Location

Hyderabad

ISSN

1098-7584

Print_ISBN

978-1-4799-0020-6

Type

conf

DOI

10.1109/FUZZ-IEEE.2013.6622582

Filename

6622582