A new paradigm for fuzzy aggregation in multisensorial image processing

Abstract

This paper presents a new paradigm for image processing in multisensorial computer vision systems based on a new interpretation of the fuzzy integral as fusion operator. Fuzzy integrals offer great chances for the implementation of the fusion stage in multisensorial systems. By exploring these possibilities a new paradigm for image processing in the framework of information fusion in multisensorial systems can be established. This new paradigm, which can be designated as Intelligent Localized Fusion (ILF), is related to Soft-Computing methodologies and the object of this paper. The performance of intelligent localized fusion operators (ILFOs), which are developed under the new introduced paradigm, are exemplary shown in the case of color edge detection in outdoor scenes. Its usage allows in this case the avoidance of false edges due to the presence of shadows. © Springer-Verlag 2001.