Multifractal analysis of three-dimensional histogram from color images

Natural images, especially color or multicomponent images, are complex information-carrying signals. To contribute to the characterization of this complexity, we investigate the possibility of multiscale organization in the colorimetric structure of natural images. This is realized by means of a multifractal analysis applied to the three-dimensional histogram from natural color images. The observed behaviors are confronted to those of reference models with known multifractal properties. We use for this purpose synthetic random images with trivial monofractal behavior, and multidimensional multiplicative cascades known for their actual multifractal behavior. The behaviors observed on natural images exhibit similarities with those of the multifractal multiplicative cascades and display the signature of elaborate multiscale organizations stemming from the histograms of natural color images. This type of characterization of colorimetric properties can be helpful to various tasks of digital image processing, as for instance modeling, classification, indexing.     

Stochastic resonance at phase noise in signal transmission

A model is developed for a periodic signal corrupted by an arbitrarily distributed phase noise and transmitted by an arbitrary memoryless system. The model establishes a new form of the phenomenon of stochastic resonance, whereby signal transmission can be enhanced by addition of noise. This is revealed by the standard signal-to-noise ratio of stochastic resonance, which here receives an explicit theoretical expression, and which is shown improvable via noise addition. This model is the first to propose a theory of stochastic resonance with phase noise. It represents a unique framework for further investigations on stochastic resonance and its applications.     

Joint acquisition-processing approach to optimize observation scales in noisy imaging

In imaging, the choice of an observation scale is conventionally settled by the operator in charge of the image acquisition, who is left alone with tuning the framing and zooming parameters of the imaging system. In a somewhat decoupled manner, the operator in charge of processing the data has access to the images after their acquisition, and seeks to extract information from the observed scene. This Letter proposes a manifestation of the interest of an alternative joint acquisition-processing approach. We demonstrate with quantitative informational measures how the choice of an observation scale can be directly related to the performance of the final information processing task. Illustrations are given with various tools from statistical information theory with possible applications of practical interest to any noisy imaging domains.     

Benefits from a speckle noise family on a coherent imaging transmission

We analyze image transmission in a coherent imaging system, in the presence of speckle noise modeled with the family of Gamma probability density functions of varying order. It is demonstrated that speckle noise can improve the transmission of a coherent image. Exact analytical expressions are obtained for both the best achievable performance and the optimal amount of speckle noise maximizing the transmission efficacy. These expressions allow us to analyze and control the conditions under which the coherent imaging system can take advantage of the speckle noise. The influence of the contrast in the coherent image, and of the order of the Gamma probability density describing the statistical fluctuations of the speckle, are given special attention. These results make a contribution to the understanding of the mechanisms of improvement by noise in nonlinear information processing.     

Exploiting the speckle noise for compressive imaging

An optical setup is proposed for the implementation of compressive sensing with coherent images. This setup specifically exploits the natural multiplicative action of speckle noise occurring with coherent light, in order to optically realize the essential step in compressive sensing which is the multiplication with known random patterns of the image to be acquired. In the test of the implementation, we specifically examine the impact of several departures, that exist in practice, from the ideal conditions of a pure multiplicative action of the speckle. In such practical realistic conditions, we assess the feasibility, performance and robustness of the optical scheme of compressive sensing.     

Signal-to-noise ratio of a dynamical saturating system: Switching from stochastic resonator to signal processor

We consider an isolated dynamical saturating system for processing a noisy sinusoidal signal, and evaluate its performance with the measure of the signal-to-noise ratio. The considered system is linear for small inputs, but exhibits saturation in its response for large inputs. This nonlinearity displays the nonlinear phenomenon of stochastic resonance for a large biased sinusoid in appropriate system parameter regions. Without the stochastic resonance phenomenon, this dynamical saturating system can achieve a signal-to-noise ratio gain exceeding unity for a noisy unbiased sinusoid. These numerical results manifest the nonlinearities and the signal-processing ability of this system acting as a stochastic resonator or a signal processor.     

An experimental study of acoustoelectric transducers with nonuniform distribution of the piezoelectric coefficient

In this paper, an experimental test of a theoretical model published previously is presented that describes the behavior of an acoustoelectric transducer with a nonuniform distribution of the piezoelectric coefficient within its bulk. Results of this theoretical model are first reviewed. Uniform and nonuniform piezoelectric transducers were fabricated, following a procedure described herein. The receive transfer functions of the transducers were recorded experimentally, and a comparison is made with the theoretical transfer functions predicted by the model, which shows good agreement. The transmit transfer functions of the uniform and nonuniform transducers were also measured and are reported. Numerical calculations of the different transfer functions given by the theoretical model for a uniform transducer associated with different backing materials are also presented, and the results are shown to be equivalent to the results following from the Mason equivalent circuit. Comparisons with experimental results and with Mason's equivalent circuit verified the new theoretical model.     

Input-output gain of collective response in an uncoupled parallel array of saturating dynamical subsystems

We explore the collective response of an uncoupled parallel array of saturating dynamical subsystems to a noisy periodic or random signal. Numerical simulation results show that a parallel array of nonlinear saturating subsystems can enhance the signal transmission via tuning the internal noise intensity and increasing the array size. The input-output gain larger than unity, described by the signal-to-noise ratio for a periodic signal or the correlation coefficient for a random signal, is observed in a form of array stochastic resonance. This stochastic resonance phenomenon can be useful for practical information-processing systems.     

Constructive action of the speckle noise in a coherent imaging system

A coherent imaging system with speckle noise is devised and analyzed. This demonstrates the possibility of improving the nonlinear transmission of a coherent image by increasing the level of the multiplicative speckle noise. This noise-assisted image transmission is a novel instance of stochastic resonance phenomena by which nonlinear signal processing benefits from a constructive action of noise.