Medical images of brain regions using graph-based method

Medical images of brain regions using graph-based method

farshad yousofli1 seyedehsan razavi2

1) Islamic Azad University,neyshabur Branch ,
2) Islamic Azad University,Mashhad Branch ,

IOI Code : XYZG-ZABAA
Publication : 2nd International Conference on Science and Engineering(2icesconf.com)
Abstract :
    Today, with the increasing proliferation of different methods of obtaining discrete information such as scanners and digital cameras with image processing has many applications. Images of this information has always been noise levels In some cases there have been problems in the image blurring of boundaries which reduces the input image is received. To solve the problem of image processing used. Image processing techniques can be considered from two areas of agreement, one aspect of hardware such as imaging, radiology and image display software and other aspects such as information extraction and processing of visual signals, such as removing noise, recovery, and the area of In medical images, especially images of the brain due to the proximity of intensity levels of brightness and edges and boundaries between tissues vague and not well realized, but in many cases for diagnosis by MRI images should be absolutely different areas of tissue To be determined The segmentation of brain images and separating the tissue where one of the most important and most basic needs in the picture Regionalization of images includes two stages: extraction of appropriate characteristics of the image and then classify these properties using an appropriate method . Zoning of low-level image processing operations aimed at partitioning the image into distinct and homogeneous areas, or the equivalent of the frontier areas. So far, many methods such as morphology binary segmentation and Markantour Mar suggested. The method presented in this paper, graph-based approach. The general framework of the method for assigning a weighted undirected graph image and segmenting the graph is based on a specific criteria. The advantage of the method is the same as the graph of the spatial information simultaneously and at high speed in the graph are made, the method for space on pixels and can be implemented with low error percentage caused zoning MRI images of the brain and improves previous methods used. This method can be used in diagnostic centers and medical imaging are considerable.

Keywords : image processing segmentation of medical images graph method