François Goudail

Influence of the noise model on level set active contour segmentation

We analyze level set implementation of region snakes based on the maximum likelihood method for different noise models that belong to the exponential family. We show that this approach can improve segmentation results in noisy images and we demonstrate that the regularization term can be efficiently determined using an information theory-based approach, i.e., the minimum description length principle. The criterion to be optimized has no free parameter to be tuned by the user and the obtained segmentation technique is adapted to nonsimply connected objects.

Target detection with a liquid crystal-based passive Stokes polarimeter

We present an imaging system that measures the polarimetric state of the light coming from each point of a scene. This system, which determines the four components of the Stokes vector at each spatial location, is based on a liquid-crystal polarization modulator, which makes it possible to acquire four-dimensional Stokes parameter images at a standard video rate. We show that using such polarimetric images instead of simple intensity images can improve target detection and segmentation performance.

Bhattacharyya distance as a contrast parameter for statistical processing of noisy optical images

In many imaging applications, the measured optical images are perturbed by strong fluctuations or boise. This can be the case, for example, for coherent-active or low-flux imagery. In such cases, the noise is not Gaussian additive and the definition of a contrast parameter between two regions in the image is not always a straightforward task. We show that for noncorrelated noise, the Bhattacharyya distance can be an efficient candidate for contrast definition when one uses statistical algorithms for detection, location, or segmentation. We demonstrate with numerical simulations that different images with the same Bhattacharyya distance lead to equivalent values of the performance criterion for a large number of probability laws. The Bhattacharyya distance can thus be used to compare different noisy situations and to simplify the analysis and the specification of optical imaging systems.

Noise minimization and equalization for Stokes polarimeters in the presence of signal-dependent Poisson shot noise

We address the optimization of Stokes polarimeters in the presence of signal-dependent shot noise, which is the dominant type of noise in certain imaging systems. We show that in some precise sense, the polarimeters optimal for additive noise are also optimal for such noise and propose polarimeter architectures in which noise variances are equalized and independent of the input polarization state.

Optimization of the contrast in polarimetric scalar images

We consider active polarimetric imaging systems that illuminate a scene with an incident polarization state and project the backscattered light on another polarization state in order to produce a scalar intensity image. We present and analyze a method for determining the configuration of illumination and analysis polarization states that maximizes the observed contrast between a target and the background when the scene is partially depolarizing and in the presence of additive Gaussian detection noise.

Near-infrared active polarimetric and multispectral laboratory demonstrator for target detection

We report on the design and exploitation of a real-field laboratory demonstrator combining active polarimetric and multispectral functions. Its building blocks, including a multiwavelength pulsed optical parametric oscillator at the emission side and a hyperspectral imager with polarimetric capability at the reception side, are described. The results obtained with this demonstrator are illustrated on some examples and discussed. In particular it is found that good detection performances rely on joint use of intensity and polarimetric images, with these images exhibiting complementary signatures in most cases.

Estimation precision of the degree of linear polarization and of the angle of polarization in the presence of different sources of noise

We consider imaging systems that measure the three first elements of the Stokes vector and deduce from them the degree of linear polarization and the angle of polarization. They require the acquisition of at least three intensity measurements, but performing more measurements is often thought to improve the estimation precision. We show that if the total acquisition time is fixed, the optimal number of measurements depends on the type of noise that affects the image: the estimation variance increases with the number of measurements N when the noise is additive; it is independent of N in the presence of Poisson shot noise and decreases with N when the angles of the analyzers fluctuate. In general, the optimal number of measurements results from a compromise on the robustness of these different types of perturbations.

Increase in depth of field taking into account deconvolution by optimization of pupil mask

We consider optimization of hybrid imaging systems including a phase mask for enhancing the depth of field and a digital deconvolution step. We propose an image quality criterion that takes into account the variability of the systems point-spread function along the expected defocus range and the noise enhancement induced by deconvolution. Considering the classical cubic phase mask as an example, we show that the optimization of this criterion may lead to filter parameters that are significantly different from those usually proposed to ensure the strict invariance of the PSF.

Spectropolarimetry of cancer change of biotissues

This work is directed to the investigation of the scope of the technique of laser polarimetry and polarization spectrometry of oncological changes of the human prostate tissue under the conditions of multiple scattering. It was shown that the third statistic moment in the intensity distribution proved to be the most sensitive to pathological changes in orientation structure. Its value in the intensity distribution of polarization image I (0 - 90) of oncologically changed tissue is 21 times higher if compared with the similar statistic parameter of the intensity distribution of the healthy tissue. The results of studies of size linear dichroism prostate gland, as healthy and affected by malignant tumor at different stages of its development was presented. Significant difference in the values of linear dichroism and its spectral dependence in the spectral range λ = 280 - 840 nm as between research facilities, and between biotissues - healthy (or affected by benign tumors) and cancer patients was shown. These results may have diagnostic value for detection and assessment of the development of cancer.

Kullback relative entropy and characterization of partially polarized optical waves

Different properties of partially polarized light are discussed using the Kullback relative entropy, which provides a physically meaningful measure of proximity between probability density functions (PDFs). For optical waves with a Gaussian PDF, the standard degree of polarization is a simple function of the Kullback relative entropy between the considered optical light and a totally depolarized light of the same intensity. It is shown that the Kullback relative entropies between different PDFs allow one to define other properties such as a degree of anisotropy and a degree of non-Gaussianity. It is also demonstrated that, in dimension three, the Kullback relative entropy between a partially polarized light and a totally depolarized light can lead to natural definitions of two degrees of polarization needed to characterize the polarization state. These analyses enlighten the physical meaning of partial polarization of light waves in terms of a measure of disorder provided by the Shannon entropy.