Geometrical interpretation of fuzzy data and shape detection

In this paper, we present a method that can be used in detecting parameters or shapes from a set of given data. Each datum is interpreted as fuzzy in the geometrical sense. This method enables us to detect more than one shape from a set of data. It also reduces the high memory requirement problem which is one of the disadvantages of the Hough transform. The problem is formulated as an optimization problem using a simple genetic algorithm. Even though this method can be applied, conceptually, to any nonlinear problem, we only show simple linear cases. Conceptual derivation and experimental results of our method are shown in this paper.