Results 1  10
of
145
Color Image Restoration Using Robust NeuroFuzzy Network Approach
"... AbstractA combination of a simple noise impulse detector and a robust neurofuzzy (RNF) network yields an effective filter for color image that corrupted by impulse noise. This method consists of 2 mode of operation, i.e. training and testing (filtering). During training, the detector impulse is em ..."
Abstract
 Add to MetaCart
. First is the use of an effective yet efficient simple impulse detector in detecting noisy pixels in the color image (feature selection). Second is the use of a memberhip function in the adaptive RNF network design, which makes the network robust to impulse noise. Experimental result shows
Comparison of Fuzzy Function Approximators
 Proc. 6 th Biennal Baltic Electronic Conference, Tallinn
, 1998
"... This paper describes three universal fuzzy function approximators  fuzzy template modeling, ANFIS and NEFPROX and compares them on the basis of their capability to model a fuzzy system. ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
This paper describes three universal fuzzy function approximators  fuzzy template modeling, ANFIS and NEFPROX and compares them on the basis of their capability to model a fuzzy system.
Some Solutions to the Missing Feature Problem in Vision
, 1993
"... In visual processing the ability to deal with missing and noisy information is crucial. Occlusions and unreliable feature detectors often lead to situations where little or no direct information about features is available. However the available information is usually sufficient to highly constrain ..."
Abstract

Cited by 53 (9 self)
 Add to MetaCart
form approximations to the Bayesian solution using Gaussian basis function networks. Simulations on a complex task (3D hand gesture recognition) validate the theory. When both integration and weighting by input densities are used, performance decreases gracefully with the number of missing or noisy features
Learning Nonparametric Kernel Matrices from Pairwise Constraints
"... Many kernel learning methods have to assume parametric forms for the target kernel functions, which significantly limits the capability of kernels in fitting diverse patterns. Some kernel learning methods assume the target kernel matrix to be a linear combination of parametric kernel matrices. This ..."
Abstract

Cited by 39 (17 self)
 Add to MetaCart
Many kernel learning methods have to assume parametric forms for the target kernel functions, which significantly limits the capability of kernels in fitting diverse patterns. Some kernel learning methods assume the target kernel matrix to be a linear combination of parametric kernel matrices
DLLite: Practical Reasoning for Rich DLs
 IN PROC. DL 2004, VOLUME 104 OF CEUR WORKSHOP PROCEEDINGS
, 2004
"... In this paper we study a DL rich enough to express UML class diagrams including ISA and disjointness between classes (but not covering constraints), typing of associations, and participation and functional cardinality constraints. For such a ..."
Abstract

Cited by 30 (6 self)
 Add to MetaCart
In this paper we study a DL rich enough to express UML class diagrams including ISA and disjointness between classes (but not covering constraints), typing of associations, and participation and functional cardinality constraints. For such a
From experiment design to closedloop control
, 2005
"... The links between identification and control are examined. The main trends in this research area are summarized, with particular focus on the design of low complexity controllers from a statistical perspective. It is argued that a guiding principle should be to model as well as possible before any m ..."
Abstract

Cited by 43 (1 self)
 Add to MetaCart
The links between identification and control are examined. The main trends in this research area are summarized, with particular focus on the design of low complexity controllers from a statistical perspective. It is argued that a guiding principle should be to model as well as possible before any model or controller simplifications are made as this ensures the best statistical accuracy. This does not necessarily mean that a fullorder model always is necessary as well designed experiments allow for restricted complexity models to be nearoptimal. Experiment design can therefore be seen as the key to successful applications. For this reason, particular attention is given to the interaction between experimental constraints and performance specifications.
Closed subgroups in proV topologies and the extension problem for inverse automata
 INT. J. ALGEBRA COMPUT
, 1999
"... We relate the problem of computing the closure of a finitely generated subgroup of the free group in the proV topology, where V is a pseudovariety of finite groups, with an extension problem for inverse automata which can be stated as follows: given partial onetoone maps on a finite set, can they ..."
Abstract

Cited by 41 (7 self)
 Add to MetaCart
We relate the problem of computing the closure of a finitely generated subgroup of the free group in the proV topology, where V is a pseudovariety of finite groups, with an extension problem for inverse automata which can be stated as follows: given partial onetoone maps on a finite set, can they be extended into permutations generating a group in V? The two problems are equivalent when V is extensionclosed. Turning to practical computations, we modify Ribes and Zalesski i's algorithm to compute the prop closure of a finitely generated subgroup of the free group in polynomial time, and to effectively compute its pronilpotent closure. Finally, we apply our results to a problem in finite monoid theory, the membership problem in pseudovarieties of inverse monoids which are Mal'cev products of semilattices and a pseudovariety of groups.
The complexity of deciding if a boolean function can be computed by circuits over a restricted basis
, 2007
"... Abstract. We study the complexity of the following algorithmic problem: Given a Boolean function f and a finite set of Boolean functions B, decide if there is a circuit with basis B that computes f. We show that if both f and all functions in B are given by their truthtable, the problem is in quasi ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
Abstract. We study the complexity of the following algorithmic problem: Given a Boolean function f and a finite set of Boolean functions B, decide if there is a circuit with basis B that computes f. We show that if both f and all functions in B are given by their truthtable, the problem
Results 1  10
of
145