Morphological shape decomposition using shape characteristics

In this paper, the morphological shape decompositions using shape directivity and shape function are proposed for the purpose of pattern recognition and image noise compression. To consider the horizontal and vertical directivity of shape, we use the ellipsoid and rectangle as primitive elements in the method using shape function, a structuring element with geometrical characteristics more similar to the shape function is preselected. The shape is decomposed into the primitive elements corresponding to the structuring elements. To prove the efficiency of the proposed algorithms, we compared them with conventional algorithms with respect to description error and coding efficiency.