Hybrid adaptive wavelet-neuro-fuzzy system for chaotic time series identification

Radial Wavelet Neural Network with a Novel Self-Creating Disk-Cell-Splitting Algorithm for License Plate Character Recognition

“…First one is to generate separable wavelets by the tensor product of several 1-D wavelet functions [19]. Another popular scheme is to choose the wavelets to be some radial functions in which the Euclidian norms of the input variables are used as the inputs of single-dimensional wavelets [21][22][23]. In this paper, the radial wavelet neural network (RWNN) is used as the means for plate character data classification.…”

Adaptive motion/force control strategy for non-holonomic mobile manipulator robot using recurrent fuzzy wavelet neural networks

“…In recent years, Fuzzy WNNs (FWNNs) have been presented in some application areas (Ho et al, 2001;Abiyev and Kaynak, 2008;Yilmaz and Oysal, 2010;Lu, 2011;Hsu, 2011;Davanipoor et al, 2012;Bodyanskiy and Vynokurova, 2013). The FWNNs, the combination of fuzzy concept and the WNNs, can bring the low level learning and good computational capability of the WNNs into fuzzy system and also high humanlike IF-THEN rule thinking and reasoning of fuzzy system into the WNNs.…”

Best basis for joint representation: The median of marginal best bases for low cost information exchanges in distributed signal representation

“…For instance, in [12] a statistical model associating fuzzy regression, nearest neighbor matching, and neural networks has been proposed for predicting the demand of natural gas by using heterogeneous rooftop unit wireless sensors; in [13] a fuzzy multi-sensor data fusion and a fuzzy Kalman feedback are used for fault detection and eective risk reduction for an integrated vehicle health maintenance system; the analysis of a neuro-fuzzy system involving adaptive wavelet activation that depends on the input signal characteristics is described in [3]; the authors of [10] show that genetic algorithms based on lifting (and thus adaptive) wavelet transforms enables relevant source separation for wide band signals while diminishing dierent types of noises; in [16] the correlation structure is used to improve the estimation accuracy of highly correlated measurements performed in multi-sensor systems.…”

“…In contrast to [3] and [10], this paper does not involve adaptive prediction and updating of wavelet coecients (wavelet lifting). We rst derive a nite set of relevant wavelet base for representing a distributed set of signals through a`lower' and`upper' wavelet basis, delimiting the set of bases-of-interest (fuzzy functional set).…”

“…In this respect and in order to approach a locally optimal solution, some references such as [1] have investigated semi-collaborative distributions schemes consisting of recursive implementations of the Karhunen-Lo eve transform. On the other hand, [3], [10], [14,4,15] have considered wavelet node splitting subject to prediction and updating stages and conditionally with respect to pre-specied collaboration schemes.…”