Knowledge Based Brain Tumor Segmentation using Local Maxima and Local Minima

In this work, we have a proposed an automatic method to brain tumor segmentation using magnetic resonance imaging (MRI) histogram. In our proposed method input is taken from abnormal slice of the MRI volume. Based on image histogram of the abnormal slice, our algorithm automatically detected the local minima and maxima using histogram smoothing techniques. Threshold value obtained from local minima between the two local maxima and segments the tumor region in T2-W MRI. These proposed works also compared with traditional clustering techniques are K-Means and FCM (Fuzzy C Means). The method yields high segmentation accuracy on various qualitative parameters and taken less computation time.