Denis de Brucq

A Colour Image Processing Model for Melanoma Detection

In this paper, we propose a method dedicated to classification between benign and malignant lesions in Dermatology in the aim to help the clinicians for melanoma diagnosis.

Identical foundation of probability theory and fuzzy set theory

Information fusion introduces special operators o in probability theory and fuzzy theory. Some serious data certify in each case these two quite distinct techniques. The article shows that four postulates are the unique aim of these two theories. Evidence theory and fuzzy set theory often replace probabilities in medicine, economy and control. Fuzzy theory is used for example in a Japanese photographic engine. We solved the challenge of unifying such different techniques. With the four postulates: noncontradiction, continuity, universality, context dependence, we obtain the same functional equation from which are deduced probability and fuzzy set theories. The same postulates apply to confidences either in the dependence or independence situation. The foundation for the various modern theories of information fusion has been unified in the framework of uncertainty by deductions. The independence between elementary confidences do not need to be understood in the sense of probabilistic meaning.

Knowledge modeling methods in the framework of evidence theory: an experimental comparison for melanoma detection

The Dempster-Shafer theory, or evidence theory, is used in different fields such as data fusion, regression or classification. Within the framework of this theory, uncertain and imprecise data are represented using belief functions. Data fusion operators as well as the decision rule of this theory were largely developed and formalized. The aim of the paper is to present modeling methods of knowledge for the initialization of belief functions. Moreover, an experimental comparison of these different modeling methods on real data extracted from images of dermatological lesions is presented.

Color image segmentation using Dempster-Shafer's theory

In this paper, we propose a color image segmentation method based on the Dempster-Shafers theory. The tristimuli R, G and B are considered as three independent information sources which can be very limited or weak. The basic idea consists in modeling the color information in order to have the features of each region in the image. This model, obtained on training sets extracted from the intensity, allows to reduce the classification errors concerning each pixel of the image. The proposed segmentation algorithm has been applied to synthetic and biomedical images in order to illustrate the methodology.