Automatic MRI Image Segmentation using Water Flow like Algorithm and Fuzzy Entropy

In this paper a new method is presented to improve the results of segmentation efficiency and accuracy. Uncertainty is an important issue in many aspects of image processing; and it must be handled in confrontation with noises and interpretative ambiguities during high level processing. Image fuzzification is one of the methods in this issue, and fuzzy entropy is the other benefits of fuzzy systems to increase reliability of segmentation. In order to provide an accurate method metaheuristic algorithms have effective results and used in numerous papers. In this paper a newer method is used, that is water flow like algorithm. The method proposed in this study can identify brain tumor fast. Input is a set of MRI slices of the patients, and output is cuts of the parts of the brain that contains a tumor that surrounded by a polygon. This method requires no image registration and it is an unsupervised technique.

Keywords:

Image Threshold , Image segmentation , water flow like algorithm , classification , Fuzzy membership function