Paper Details: Downloads: 775
Serial Number: P1150812004
Title: An Improved Iterative Morphological Decomposition Approach for Image Skeletonization
Authors: V. Vijaya Kumar and A. Srikrishna and D. V. L. N. Somayajulu and B. Raveendra Babu
Abstract: The morphological skeleton method is a popular approach for morphological shape representation. The present paper proposes a new algorithm for skeletonization of 2D images, based on primitive concepts of morphology which is an extension of the morphological binary skeleton. The proposed paper utilizes structuring elements of size n×n, where n=2, 3, 4...m. A fine comparison is made on the effect of skeletonization based on the size of structuring element and their repetitiveness. One of the major problems of existing approaches is that at what iterative step the skeletonization process should stop. This problem is solved in the present paper by experimenting the existing and proposed method on various images with different structuring elements. One of the problems of multi-resolution (scale-space) approach to skeleton construction is that they are sensitive to boundary noise. This paper advocates this problem as well. The present method is experimented on various shapes, alphabets and digits. Good results are obtained and they are compared with the other existing techniques.
Keywords: Mathematical morphology, Structuring elements, Shape representation, Skeletonization
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 8
Issue: 1
Submission Date: 3/1/2008 12:00:00 AM
Review Date: 4/1/2008 12:00:00 AM
Publishing Date: 6/1/2008 12:00:00 AM
Article Downloads: 775
Download:

Facebook