Top-Down Method Used for Pancreas Segmentation
Pradip M. Paithane1, S. N. Kakarwal2, D. V. Kurmude3
1Pradip M. Paithane*, Computer Engineering, Dr. BAMU, Aurangabad, India.
2Dr. S.N. Kakarwal, Professor and Head of Department, Computer Engineering, PES Engineering College, Aurangabad, India.
3Dr. D.V. Kurmude, Professor, and Head of Department, Physics Department, Milind College, Aurangabad, India.
Manuscript received on December 16, 2019. | Revised Manuscript received on December 22, 2019. | Manuscript published on January 10, 2020. | PP: 1790-1793 | Volume-9 Issue-3, January 2020. | Retrieval Number: B7422129219/2020©BEIESP | DOI: 10.35940/ijitee.B7422.019320
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Image segmentation is actively an imperative title role in image analysis. Image segmentation is advantageous in many applications like traffic detection, surface crack identification, medical image analysis, face recognition, crop disease detection. Two Approaches are used for automatic pancreas segmentation. Top-Down and Bottom-Up approach used for CT image segmentation. In Top Down approach, Grey Level Co-occurrence Matrix, Simple Linear Iterative Clustering, Scale-Invariant Feature transform, Novel Modified Kernel fuzzy c-means clustering (NMKFCM) and Kernel Density Estimator methods used and automatic bottomup technique is used for pancreas subgrouping in C.T. scans. Top-Dow approach accuracy rate is less than bottom-up approach. Top-down approach required less time period as compare to bottom-up approach. In top-down approach, input image manually selected and processed it. KDE, NMKFCM and SIFT are used to detect feature of image. NMKFCM works on neighborhood point value. In KDE, Edge detection based on the kernel estimation of the probability density function .In SIFT, comprehensive information of local feature of image is focused.
Keywords: Abdominal Computed Tomography (CT),Clustering, GLCM, Kernel Density Estimator, Scale Invariant Transform(SIFT),NMKFCM.
Scope of the Article: Clustering