No Detail Found!

Adjustment of normalized cut parameters using neural networks

Author(s)
SAMIA ELSIMARIE Information Systems Department, Faculty of Computers and Information, Suez Canal University, Egypt
BENBELLA SAYED Information Systems Department, Faculty of Computers and Information, Suez Canal University, Egypt
AHMED OTHMAN Information Systems Department, Faculty of Computers and Information, Suez Canal University, Egypt
GHADA ELTAWEEL Computer science Department, Faculty of Computers and Information, Suez Canal University, Egypt
Abstract
Image segmentation is an important process that used in both quantitative and qualitative analysis of medical ultrasound images, but medical images have features of strong noise and poor contrast and the results of image segmentation may not be good with traditional segmentation methods. in this paper we segment breast ultrasound medical images based on texture features and graph cut ,gray- level spatial dependence matrix(GLSDM) used to extract texture feature parameters ,the similarities matrix is created according to the parameters of texture feature and gray intensity of pixels .normalized cut spectral graph theoretic framework used to segment image depending on the similarity matrix. This paper introduces a new approach to overcome the problems associated with medical image segmentation such that the proposed approach (Neural Normalized Cut) has the ability to adjust the parameters of normalized cut segmentation technique , Neural normalized Cut has applied for breast ultrasound images , the results show the ability of neural normalized cut to adjust multiple parameters and enhance image segmentation ,especially for medical images.
Publication Details
Page(s) 1290-1300
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 98, Issue: 8, Year: 2020

Request the full text from PASTIC

Keywords
K-Means, Genetic algorithms, Surf, thresholding, Neural Networks, Graph Cut, Normalized cut, image segmentation, Neural Normalized Cut Segmentation, Medical Ultrasound Images, Gray- Level Spatial Dependence Matrix (GLSDM)
0

Citations

0

Downloads

145

Views