Gray-level Morphological Operations for Image Segmentation and Tracking Edges on Medical Applications

摘要

In this paper we mainly focused on grayscale images and how these images can be decomposed into number of peaks. This decomposition is called the peak analysis of grayscale images. Due its sensitive definition, there is a wide range of applications for peak analysis. The detection criterion expresses the fact that important edges should not be missed. It is of paramount important to preserve, uncover or detect the geometric structure of image objects. Thus morphological filters, which are more suitable than linear filters for shape analysis, play a major role for geometry based enhancement and detection.rnA new method for image segmentation and tracking edges based on morphological transformation is proposed. This algorithm uses the morphological transformations dilation and erosion. A gradient determined grey level morphological procedure for edge increase and decrease is present. First, the maximum gradient, in the local neighborhood, forms the contribution to the erosion of the center pixel of that neighborhood. The gradients of the transformed image are then used as contributions to subsequent dilation of eroded image. The edge sharpening algorithm is applied on various sample images. Proposed algorithm segments the image by preserving important edges.