Fuzzy Interpretation of Image Data
Joon Hee Han, Tae Y. Kim, László Tamás Kóczy · Studies in fuzziness and soft computing · 2000
In image formation, we usually consider two things: the intensity and the location of a pixel. Because of several reasons, there could be uncertainty in the image brightness and also in the location of a pixel. We consider the cases where the observed image data or entities computed from it are inherently fuzzy. Based on this idea, we have considered shape detection methods and representation of edges using fuzzy set theory. In shape detection method, an image point is considered as a fuzzy data. By combining this concept and the Hough transform algorithm, a fuzzy Hough transform algorithm is introduced. More general shape detection paradigm is given by defining the cardinality of a shape. Edges, which is one of the basic entities of an image, is generalized using the fuzzy set theory. An edge evaluation criteria for the edges are also presented. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.