Brain Tumar Detection

Abstract
In this project our main objective is to detect the tumor and calculate its area based on classifying the images. Death rate among people has been increased due to diseases like brain tumor. The brain tumor that can be identified by using image processing. The proposed system enables automatic detection of brain tumor through MRI. So our main objective is to study, analyze and enhance MRI Image from the existing algothrim. Certain traditional approach requires manually extracting the location where the tumor is build.
Now a day’s automatic approach is there to extract the tumor from the automatic machines where the doctor can easily find out the tumor from the machine and also give the report of the patient on the spot without wasting the time. It helps to save lives of many patients who are suffering from brain tumor. Now a day’s system is very fast doctors can cure the disease of patients quickly by using automatic machines. The system also minimizes human intervention to reduce human errors while keeping the accuracy offered by fully automated approach. This project is very beneficial in terms of detecting the tumor by applying preprocessing, segmentation and classification part with dataset of seventy (70) images. The Doctor can save a lot of time to solve this problem by detecting the tumor at an exact location by performing the best methods. Today Doctors are using developed system to find the accurate location of the tumor and immediately find the results and proceed to the treatment of the patient.

i

LIST OF FIGURES

FIG

FIGURE DETAILS

NO

PAGE
NO

3.1

Proposed Methodology Of Brain Tumor Detection

12

4.1

Flowchart of Segmentation

20

4.2

Flowchart of Feature Extraction

21

4.3

Flowchart of Morphological Operators

22

4.4

Flowchart of Classification

23

5.1

Input Image

27

5.2

Pre-processing Image

27

5.3

Threshold Segmentation

28

5.4

K-means