Modelling by fuzzy approach uncertainties in image analysis

Abstract

Investigating some objects in all areas, they are mapped to the digital raster image using different kinds of sensors, the obtained image is only an approximation to the real-world object. Due to imperfections in either the image data or the edge detector, there may be missing points or pixels on lines as well as spatial deviations between ideal line and the set of imprecise points obtained from the edge detector. The overall effect is an image that has some distortion in its geometry. In this paper it is presented a mathematical model based on fuzzy sets. In this way there covered these uncertainties, and it is obtained correct interpretation and important decisions in different important areas as image analysis (imprecise feature extraction), GIS (imprecise spatial object modelling), robotics (environment models), and in medicine (DICOM medical images).