Binary Morphological Model in Refining Local Fitting Active Contour in Segmenting Weak/Missing Edges
2012 International Conference on Advanced Computer Science Applications and Technologies (ACSAT) (2012)
Nov. 26, 2012 to Nov. 28, 2012
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ACSAT.2012.60
Medical images are known to have poor quality which leads to difficulty in vision and segmentation process. Mainly, noise and intensity in homogeneity are two main characteristics that lead to gaps (missing at edges) at the boundary of the desired object. This paper investigated method that managed to smooth the image texture in order to overcome the gaps problem. Our method adopts the morphological closing operations using the diamond-shape structuring elements to overcome the above-mentioned problem. We applied the dilation and erosion operation to expand and later smooth the regions with gaps. Our method shows satisfaction results when dealing with binary image rather than working with gradient. The results obtained shows better accuracy as the evolving curve is following the pixels value in the binary image. The method proposed is executed based on the output from Local binary fitting energy.
computer vision, edge detection, image segmentation, image texture, medical image processing
N. Kamaruddin, H. A. Jalab, R. Zainuddin and N. A. Abdullah, "Binary Morphological Model in Refining Local Fitting Active Contour in Segmenting Weak/Missing Edges," 2012 International Conference on Advanced Computer Science Applications and Technologies (ACSAT), Kuala Lumpur, 2013, pp. 446-451.