Research On Segmentation Techniques
Sushma B M1, Akshatha Prabhu2
1Sushma B M a student of Amrita School of Sciences, Mysuru. BCA graduate from SDM college, Mysuru University (Karnataka), India.
2Ms. Akshatha Prabhu, Assistant Professor Department of Computer Science.
Manuscript received on 18 June 2019 | Revised Manuscript received on 25 June 2019 | Manuscript published on 30 June 2019 | PP: 1783-1786 | Volume-8 Issue-5, June 2019 | Retrieval Number: E7605068519/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Brain cancer is the most dangerous disease in the world and serious leading deaths every day. Early recognition and treatment can save the cancer affected human’s life. Identification of brain cells has been a challenging task in medical field. The size and shape of tumor in the brain are different for each patient. Brain is one of the important organs. Brain tumor is the mass or growth of abnormal cells in the human body. Brain tumor can be cancerous (malignant) or noncancerous (benign). Cancerous tumor might be primary or secondary. Primary brain tumors are tumors where the cancer first grows in the body. Secondary tumors are the tumors where primary cancer spreads to another part of the body. Magnetic Resonance Imaging (MRI) gives the complete images of the human body. MRI technique generates the images using strong magnetic fields and radio waves. MRI scanning technique is the current best scanning technique in the health care industry, MRI images can be used by the doctors to identify the tumor part in the brain. MRI scanning technique is best because it does not use x-rays. Preprocessing step can be used to better the quality of the image, to improve the image, to remove the noise. The goal of the segmentation is to analyze the image, it changes the illustration of the image that is significant and easier to examine. Comparison has been made on Otsu threshold, k means and texture filtering segmentation techniques. Different techniques provide different result. The best and efficient method can be identified by measuring AROC curve.
Keywords: Preprocessing, Image Segmentation, Otsu, K Means, Texture Filtering, AROC Curve, Specificity, Sensibility.
Scope of the Article: Image Processing