Author Email: Virupakshi.firstname.lastname@example.org
In this paper an improved approach for segmentation of brain tumor using MRI images has been presented. This approach consists of three steps, namely: preprocessing, segmentation and post processing. The preprocessing has been carried out using skull stripping and histogram equalization techniques. The region of interest is segmented by applying conventional Level Set method to the preprocessed image. The segmented image consists of distortions at the boundaries which lead to boundary leakage problem. This can be minimized by applying dynamic Thresholding technique and region localization method. The proposed method achieved better accuracy of segmentation as compared to the conventional level set segmentation method.
Image Segmentation, MRI sample, Level Set coding, Region Detection of MRI sample.
Image segmentation is an important task of image processing. Properly segmented image provides the basis of further very high level image understanding, subsequent planning for diagnosis. Through manual clustering by human experts is a tedious task due to very slight variations, similarity among affected and original biological tissues and larger data set for analysis. These problems go more complicated in general and brain tumor in particular. A brain tumor is a disease in which tissues grow irresistibly in the brain. Brain tumors are classified into two types namely: benign and malignant. In the case of benign tumors, they are incapable of unfolding outside the brain and their growth is self-limited. Due to their location sometimes tumors creates problem. They are curable using radiation or surgery method. In the case of malignant tumors, they are called brain tumors; these tumors can spread not only in the brain as well as other parts of the body at faster rate. Malignant brain tumors create problem if it is not treated.
We have presented an improvised brain tumor recognition system for the MR images using the level set coding and dynamic Thresholding binarization logic. It is found that the system proved in better segmentation during the tumor recognition process. The accuracy level and iteration time is found to be improved considerably in tumor recognition compared to the conventional LS segmentation. Since implementation of proposed method is done on single image so no worries of variance in intensity across multiple MR images. This method is suitable for 2-D MRI images; in future we will try to extend it for 3-D MRI images
256 total views, 1 views today