%0 Journal Article %1 Ng2008 %A Ng, H. P. %A Huang, S. %A Ong, S. H. %A Foong, K. C. %A Goh, P. S. %A Nowinski, W. L. %D 2008 %J Conf Proc IEEE Eng Med Biol Soc %K Algorithms; Analysis; Anatomic; Automated, Cluster Computer-Assisted, Computer-Assisted; Diagnostic Distribution; Humans; Image Imaging, Models, Normal Pattern Processing, Recognition, Reproducibility Results; Signal Software Statistical; Theoretical; methods; of %P 4039--4042 %T Medical image segmentation using watershed segmentation with texture-based region merging. %V 2008 %X The use of the watershed algorithm for image segmentation is widespread because it is able to produce a complete division of the image. However, it is susceptible to over-segmentation and in medical image segmentation, this meant that that we do not have good representations of the anatomy. We address this issue by thresholding the gradient magnitude image and performing post-segmentation merging on the initial segmentation map. The automated thresholding technique is based on the histogram of the gradient magnitude map while the post-segmentation merging is based on the similarity in textural features (namely angular second moment, contrast, entropy and inverse difference moment) belonging to two neighboring partitions. When applied to the segmentation of various facial anatomical structures from magnetic resonance (MR) images, the proposed method achieved an overlap index of 92.6\% compared to manual contour tracings. It is able to merge more than 80\% of the initial partitions, which indicates that a large amount of over-segmentation has been reduced. Results produced using watershed algorithm with and without the proposed and proposed post-segmentation merging are presented for comparisons.

@article{Ng2008, abstract = {The use of the watershed algorithm for image segmentation is widespread because it is able to produce a complete division of the image. However, it is susceptible to over-segmentation and in medical image segmentation, this meant that that we do not have good representations of the anatomy. We address this issue by thresholding the gradient magnitude image and performing post-segmentation merging on the initial segmentation map. The automated thresholding technique is based on the histogram of the gradient magnitude map while the post-segmentation merging is based on the similarity in textural features (namely angular second moment, contrast, entropy and inverse difference moment) belonging to two neighboring partitions. When applied to the segmentation of various facial anatomical structures from magnetic resonance (MR) images, the proposed method achieved an overlap index of 92.6\% compared to manual contour tracings. It is able to merge more than 80\% of the initial partitions, which indicates that a large amount of over-segmentation has been reduced. Results produced using watershed algorithm with and without the proposed and proposed post-segmentation merging are presented for comparisons.}, added-at = {2011-03-11T12:21:24.000+0100}, author = {Ng, H. P. and Huang, S. and Ong, S. H. and Foong, K. C. and Goh, P. S. and Nowinski, W. L.}, biburl = {https://www.bibsonomy.org/bibtex/2c95eb0062f50e78f1c2fcf64fc8a8357/jmaiora}, institution = {Biomedical Imaging Lab, Agency for Science Technology and Research, Singapore. ng_hsiao_piau@sbic.a-star.edu.sg}, interhash = {286c3a669f09dab55c75edd42cd0622c}, intrahash = {c95eb0062f50e78f1c2fcf64fc8a8357}, journal = {Conf Proc IEEE Eng Med Biol Soc}, keywords = {Algorithms; Analysis; Anatomic; Automated, Cluster Computer-Assisted, Computer-Assisted; Diagnostic Distribution; Humans; Image Imaging, Models, Normal Pattern Processing, Recognition, Reproducibility Results; Signal Software Statistical; Theoretical; methods; of}, language = {eng}, medline-pst = {ppublish}, owner = {Josu}, pages = {4039--4042}, pmid = {19163599}, timestamp = {2011-03-11T12:21:26.000+0100}, title = {Medical image segmentation using watershed segmentation with texture-based region merging.}, volume = 2008, year = 2008 }