SEGMENTATION AND DETERMINATION OF BRAIN TUMOR BY BOUNDING BOX METHOD
An Intracranial Neoplasm (Brain Tumor) occurs when abnormal cells form within the brain. There are two main types of tumors: Malignant (Cancerous Tumors) and Benign tumors. Cancerous or non-cancerous mass and growth of abnormal cells in the brain leads to the formation of brain tumor. In order to reduce the increasing fatality rate caused by brain tumor, it is necessary to detect and cure the affected region early and efficiently. Initially, pre-processing is performed, in this phase image is enhanced in the way that finer details are improved and noise is removed from the image. During pre-processing, filters are applied on an input grey scale image to remove unwanted impurities. Filtered image thus obtained is free from impurities. Processing of an image is performed next. Image segmentation is based on the division of the image into regions. Division is done on the basis of similar attributes. Post processing is done using threshold and watershed segmentation. During post processing, the filtered image is forwarded for threshold segmentation along with SVM classifier. Threshold segmentation usually transforms the image in a binary format based on a threshold value. SVM analyze data for classification and regression analysis. Watershed segmentation groups the pixels of image based on their intensities. Morphological operations are applied to the converted image. Boundary extraction is a major part of research which uses fast bounding box algorithm which detects the affected area in motion.