A Watershed Segmentation Process based on Progressive Median Filtering & Gradient Map
Ankur Chourasia1, Akhilesh Singh Thakur2, Vibha Tiwari3
1Ankur Chourasia, EC Department, GGITM Bhopal, India.
2Mr. Akhilesh Singh Thakur , EC Department, GGITM Bhopal, India.
3Dr. Vibha Tiwari, EC Department, GGITM Bhopal, India.
Manuscript received on July 25, 2013. | Revised Manuscript received on August 07, 2013. | Manuscript published on August 30, 2013. | PP: 155-159 | Volume-2, Issue-6, August 2013. | Retrieval Number: F2009082613/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper we present a digital image segmentation algorithm that is effective and offers robustness while minimizing the over segmentation issues. The proposed algorithm is designed to use the combination of Median-filtering, soft thresholding and watershed segmentation method, and sobel gradient map was used to perform image segmentation and edge detection tasks. In brief, median filter is performed on the image to limit the problem of undesirable over-segmentation results produced by the watershed algorithm. Soft thresholding is carried based on the region’s maximum value to obtain binary segments of various classes to boast the watershed algorithm performance. The gradient map is created based on the edge strength of the image using sobel operators. In addition, the simulations results reveal that the proposed system offers improved segmentation results in comparison with the regular watershed algorithms.
Keywords: Watershed algorithm, Segmentation, Media filter, Sobel operator, Morphological operation.