We present a new method for the reduction of noise in images, using a wavelet transform. The method relies on two principles. The rst is the characterization of the local function regularity by wavelet coeecients. The second is an a priori, geometrical model for wavelet coeecients. Both are combined in a Bayesian framework, to compute for each wavelet coeecient the probability of being \suu-cie...

A new approach to Bayesian reconstruction is introduced in which the prior probability distribution is endowed with an inherent geometrical flexibility. This flexibility is achieved through a warping of the coordinate system of the prior distribution into that of the reconstruction. This warping allows various degrees of mismatch between the assumed prior distribution and the actual distributio...

We introduce a simple geometrical two-dimensional continued fraction algorithm inspired from dynamical renormalization. We prove that the algorithm is weakly convergent, and that the associated transformation admits an ergodic absolutely continuous invariant probability measure. Following Lagarias, its Lyapunov exponents are related to the approximation exponents which measure the diophantine q...

The iterative proportional fitting procedure (IPF-P) is an algorithm to compute approximately probability measures with prescribed marginals. We propose two extensions of the IPF-P, called conditional iterative proportional fitting procedures (CIPF-P), so that, additionally, given conditional distributions are taken into account. This modification is carried out by using the geometrical interpr...

We show that the correct entropy, temperature (and absorption probability) of non-extremal black p-brane can be reproduced by a certain thermodynamical model when maximizing its entropy. We argue the relation of the model and the geometrical similarity of black p-brane and R AdS black hole at near horizon region. We also argue the relation of the maximization of the entropy and a saddle point a...

Ray tracing and the Geometrical Theory of Diffraction are applied to a simulation of the radio channel experienced by equipment communicating inside buildings. Comparisons are made between the simulated results and electromagnetic measurements through the use of probability distributions fitted to the data. System parameters are determined from the simulation, and used to recreate the channel i...

Road traffic volumes in intercity roads are generally estimated by probability functions, statistical techniques or meta-heuristic approaches such as artificial neural networks. As the road traffic volumes depend on input variables and mainly road geometrical design, weather conditions, day or night time, weekend or national holidays and so on, these are also estimated by pattern recognition te...

In this paper we present a new blind separation of sources (BSS) algorithm based on second order statistics (SOS) and geometrical approaches. The new algorithm can separate multisources from their instantaneous mixtures obtained by multisensors. In the case of p sources and p sensors, the algorithm can be decomposed into p steps: First, one should transform the mixing signals to orthogonal sign...

We consider the cluster mass distribution between two lines of arbitrary orientations and lengths in porous media in three dimensions, and model the porous media by bond percolation at the percolation threshold pc. We observe that for many geometrical con.gurations the mass probability distribution presents power law behavior. We determine how the characteristic mass of the distribution scales ...