Authors : R. Latha and S. Sentil Kumar
Abstract: In this study, an algorithm using traditional morphology for the extraction of blood vessels in cardio angiographic images is compared with that of the fuzzy watershed transformation. Blood vessels usually have poor local contrast and the existing Edge Detection algorithms do not give better results. In traditional morphology, an edge detection operation based on laplacian of Gaussian is performed but in fuzzy morphology, a fuzzy morphological multigradient watershed transformation is carried out to extract the boundary of vessels. In fuzzy watershed transformation, the marker extraction of the gradient image is obtained by a thresholding technique to avoid over segmentation and given as input to the watershed transformation. Finally, a set of alternating filters are used to segment the vessels from the cardiac images in both the methods. Simulation results show that the proposed algorithm using fuzzy watershed transformation produces better segmentation result on the basis of quality and takes less execution time when compared to the algorithm using traditional morphological filters.
R. Latha and S. Sentil Kumar, 2014. Comparison of Traditional Morphology and Fuzzy Watershed Transformation in Medical Image Segmentation. International Journal of Soft Computing, 9: 194-199.