Abstract

The watershed transform is the method of choice for medical image segmentation. Separating overlapping objects in an image is one of the challenging tasks in image processing operations. Segmentation of brain MR image using marker controlled watershed transform works better if we can identify or mark foreground objects and background locations. This paper deals with a novel segmentation method giving realistic results on brain tumor boundary detection using brain MR images. This method can be easily used to diagnose tumor boundary with a faster rate and it is accurate compare to the existing methods.