Medial Axis Transformation based Skeletonzation of Image Patterns using Image Processing Techniques

Navjot Mann, Parminder Singh

Abstract: The medial axis of an image pattern is the loci of all inscribed disks that touch two or more boundary points without crossing any of the boundaries. The medial axis transform (MAT) is a powerful representation for objects with inherent symmetry or near-symmetry. The medial axis of 2-D image patterns provides a conceptual design base, with transition to a detailed design occurring when the radius function is added to the medial axis or surface. To make such a design tool practicable, however, it is essential to be able to convert from an MAT format to a boundary representation of an object. In the proposed work, the medial axis transform has been extracted using the Euclidean distance transform based computation. The image pattern u prepared initially in binary form and then distance of each non-zero pixel to its closest zeroed pixel is computed. This process continues till the entire image pattern is scanned to its core.

Keywords: Medial Axis Transform, Euclidean Distance Transform, Charge Coupled Device