A Method For Constructing Morphological Operators With Adaptive Circular Structure Elements

The traditional mathematical morphology has the fixed structuring element while processing the whole image,which is easily to result in over-processes or the below-processes to the input image,since the variation of image structures regarding e.g.shape,size,and orientation often provides a challenge when processing all points identically.Therefore,in order to introduce the mathematical morphology into the increasingly wide range of image processing applications,it is necessary to design an adaptive structure element.This paper has proposed a novel algorithm that has constructed adaptive elliptical structuring elements via estimating the local anisotropy of an image based on the nonlinear structure tensor.Firstly this algorithm analyzes the image information contained in the nonlinear structure tensor,applying the corner and edge strength defined by the eigenvalues of the nonlinear structure tensor to distinguish the different parts of the input image,such as edges,corners and smooth region et al.Then combings the requirement of the structuring element’s shape needs to satisfy the image space adaptability,the parameters of the elliptical structuring element are defined.In addition,it may improve the resolution of the elliptical structuring elements to make them close to the ideal ellipse,which force us to improve the resolution of the image simultaneously.Finally,erosion,dilation,opening,closing and Hit-or-Miss transform are redefined according to the presented structuring elements.The processed results and the quantitative analysis show that the novel morphological operators have more advantages in structure adaptation,corner protection,filtering and targets extraction than the others.