Brain Tumor Detection Using Marker Based Watershed Segmentation from Digital MR Images

 Abstract— This paper presents a method for detection of brain tumor from Magnetic Resonance Image. Pre-processing the image makes it ready for applying the watershed segmentation. Pre-processing includes image resizing, conversion to gray. Gradient magnitude is to be computed before applying the segmentation and magnitude of these gradients is computed using the sobel mask. Watershed segmentation is used for detecting the tumor. The basic watershed algorithm is well recognized as an efficient morphological segmentation tool however, a major problem with the watershed transformation is that it produces a large number of segmented regions in the image around each local minima embedded in the image. A solution to this problem is to use marker based watershed segmentation. Connected component analysis extracts the regions which are not separated by boundary after region boundaries have been detected. Finally