Steve Guyot

Texture analysis based on local analysis of the Bidimensional Empirical Mode Decomposition

Abstract.The main contribution of our approach is to apply the Hilbert-Huang Transform (which consists of two parts: (a) Empirical Mode Decomposition (EMD), and (b) the Hilbert spectral analysis) to texture analysis. The EMD is locally adaptive and suitable for analysis of non-linear or non-stationary processes. This one-dimensional decomposition technique extracts a finite number of oscillatory components or “well-behaved” AM-FM functions, called Intrinsic Mode Function (IMF), directly from the data. Firstly, we extend the EMD to 2D-data (i.e. images), the so called bidimensional EMD (BEMD), the process being called 2D-sifting process. The 2D-sifting process is performed in two steps: extrema detection by neighboring window or morphological operators and surface interpolation by radial basis functions or multigrid B-splines. Secondly, we analyse each 2D-IMF obtained by BEMD by studying local properties (amplitude, phase, isotropy and orientation) extracted from the monogenic signal of each one of them. The monogenic signal is a 2D-generalization of the analytic signal, where the Riesz Transform replaces the Hilbert Transform. The performance of this texture analysis method, using the BEMD and Riesz Transform, is demonstrated with both synthetic and natural images.

Near real-time polarimetric imaging system

Abstract. A new imaging technique which enables near real-time multispectral acquisition of the so-called degree of polarization (DOP) in polarimetry using incoherent white light is described. The experimental setup allows the interactive and dynamic acquisition of DOP for all the possible elliptic polarization states. In such a way, a complete chart of light-matter interaction can be obtained and besides many structure details can be enhanced. Thus, we present the calibration and first images acquired with this system. The particular characteristics of this setup allow it to be the perfect candidate for in vivo as well as ex vivo medical applications.

Conceiving an easy-to-understand and automated polarimeter: application to a study of collagen

In this paper, the authors describe the design and the experimentation of a polarimeter aimed at characterizing the properties of some scattering materials (e.g., a scattering liquid containing a foreign body, or skin components, such as collagen). The polarimeter presents the advantage of offering an easy understanding of Stokes and Mueller formalism. Thus, it is typically suited for university students in physics (Masters level), giving them a clear understanding of polarization and birefringence. Another application is possible in schools of engineering, where students may use their design abilities to improve the apparatus performance. Apparatus automation was performed as well, providing easier use for experiment (commodity of use, reproducibility, and possible multiplication of measurements). The apparatus was tested in transillumination and backscattering modes. More generally, measurements can be performed with angles varying from 0/spl deg/ to 150/spl deg/ between the propagation direction and the line from the sensor to the sample material. In addition, the sensor can be either a photodiode (results presented here) or a two-dimensional (2-D) sensor charge-coupled device (CCD) camera. As an application, the apparatus was used to differentiate irradiated and nonirradiated collagen.

A Mueller polarimetric imaging system for biomedical applications

We present a new polarimetric imaging system based on liquid crystal modulators, a spectrally filtered white light source and a CCD camera. The whole Mueller matrix image of the sample is measured in around 5 seconds in transmission mode. The instrument design, together with an original and easy-to-operate calibration procedure provides a high accuracy (better than 1.5% for the normalized Mueller matrix) over a wide spectral range. The data can be processed with different algorithms. Results on hepatic biopsies with different grades of fibrosis are presented.

Circular Statistics of Fractional Fields

The Riesz transform, associated with a model of circular distributions, is used for the analysis of complex structures of fractional fields through their local direction statistics

Stochastic analysis of speckle in backscattering

In this paper, we propose a new approach of the speckle statistics in backscattering imagery. Applicate the Brownian motion theory to the speckle permit us to extract stochastic parameters to characterize it. It seems more powerful than the classical frequential approach to characterize and classify speckle. We present an test application of this method on a human skin.