Spatial fuzzy C-means Clustering based Liver And Liver Tumor Segmentation on Contrast Enhanced CT Images
Sajith A.G1, Hariharan.S2
1Sajith A.G, Electrical Engineering, College of Engineering, Trivandrum, India.
2Dr. Hariharan S, Electrical Engineering, College of Engineering, Trivandrum, India.
Manuscript received on January 20, 2015. | Revised Manuscript received on February 14, 2015. | Manuscript published on February 28, 2015. | PP: 136-139 | Volume-4 Issue-3, February 2015. | Retrieval Number: C3775024315/2013©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: Analysis of CT images plays an important role in liver tumour segmentation. Segmentation methods include thresholding, region growing, splitting and merging etc. Segmentation methods are of two types fully automatic and semi-automatic. It is the first and essential step for the diagnosis of liver diseases. Region based segmentation plays an important role in CT liver image analysis. In this paper a hybrid image processing method is presented based on spatial fuzzy C means clustering combined with Mumford Shah model. In image processing Mumford shah model is used for minimizing an energy function involving a piecewise smooth representation of the image. Thus we can detect interior contours automatically enhanced the blurred contours and increase the robustness of an image with less number of iterations. Thus we can improve the segmentation of liver image thereby increasing the detection of tumour effectively.
Keywords: Spatial FCM, Mumford Shah model, Image segmentation, CT liver image analysis.