Brain Tumor Detection By Various Image Segmentation Techniques With Introducation To Non Negative Matrix Factorization

Abstract

Image segmentation is one of the fundamental approaches of digital image processing. During past few years, brain tumor segmentation in magnetic resonance imaging (MRI) has become a popular research area in the field of medical imaging system. MRI is used in radiology for analysing internal structures and makes easy to extract the required region. Thresholding is the simple approach to introduce to the morphological operations which are useful for the detection of the tumor but not all tumor can be specifically detected by this technique so region growing is another technique which provide seed point approach to the segmenter ROI region so the tumor is easily detected and also further used for the classification purpose. Nonnegative Matrix Factorization is one of the most promising technique to reduce the dimensionality of the data.NMF has been applied earlier to the image Processing methods such as Pattern analysis and Text mining and now in this paper it is mainly used as a uninterruptable decomposition approach for detection of tumor and to further classify into various types and also for feature extraction.NMF aims to find two non negative matrices whose product closely approximate the original matrix.NMF contains all matrices to contain only non negative elements and the NMF results shows no cancellations, linear super position only and considerable sparsity.