BRAIN TUMOR DETECTION USING OBJECT LABELING ALGORITHM & SVM

Abstract

In this paper, we propose an efficient brain tumor uncovering method, which can detect tumor and localize it in the brain MRI images. This method extracts the tumor by using K-means algorithm followed by Object labelling algorithm. For better result we can cluster the identified regions and apply object labelling there, also some preprocessing step (median filtering) is used for tumor detection purpose. At the end the training and testing by SVM will be perform, SVM find out the patterns of this to use in future and comparison purpose so it will enhance the efficiency of system and tumor detection will get easier task in medical field. It is expected that the experimental results of the proposed method will give better result in comparison to other existing techniques.