Title: Adaptive image restoration by a novel neuro-fuzzy approach using complex fuzzy sets

Authors: Chunshien Li

Addresses: Laboratory of Intelligent Systems and Applications, Department of Information Management, National Central University, No. 300, Jung-da Rd., Jung-li City, Taoyuan, Taiwan

Abstract: A complex neuro-fuzzy approach using new concept of complex fuzzy sets and neuro-fuzzy system is presented to deal with the problem of adaptive image noise cancelling (AINC). An image can be tainted by unknown noise, resulting in the degradation of valuable image information. A complex fuzzy set (CFS) is characterised in the unit disc of the complex plane by a complex-valued membership function that includes an amplitude function and a phase function. Based on the nature of CFSs, several CFSs can be used to design a complex neural fuzzy system (CNFS) for the application of AINC. To train the CNFS, a hybrid learning method is used, where the algorithm of artificial bee colony (ABC) and the method of recursive least squares estimator (RLSE) are integrated in a complementarily hybrid way. Three cases are used to test the proposed CNFS for image restoration. The experimental results by the proposed CNFS approach are compared with those by other approaches and the proposed approach has shown promising performance.

Keywords: complex fuzzy sets; CFS; complex neuro-fuzzy system; CNFS; artificial bee colony; ABC algorithm; recursive least squares estimator; RLSE; image restoration; fuzzy logic; neural networks; image noise cancelling; image processing.

DOI: 10.1504/IJIIDS.2013.057419

International Journal of Intelligent Information and Database Systems, 2013 Vol.7 No.6, pp.479 - 495

Accepted: 14 Jun 2013
Published online: 31 Mar 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article