Enhanced Image Filtrationusing Threshold based Anisotropic Filter for Brain Tumor Image Segmentation

Abstract

In recent days, the medical image processing plays a vital role in medical research. Image classification, filtration, and segmentation are the major processing techniques. In this paper, the brain tumor image filtration and segmentation is processed with image processing technique, which uses anisotropic filter and it is enhanced with the threshold based segmentation and bounding box method. The morphological operation of MRI brain image before performs the anisotropic diffusion filter, which reduces the contrast between consecutive pixels. This image is resized and it converted into grey scale image. Based on the threshold value, the tumor region is segmented by bounding box method. Here the morphological process is applied to gather information of tissue for obtaining the possible values. Moreover, the process is not occurred on the tumor less image or hollow image, which improves the detection process. Main goal of this method is to obtain the 3D image tissue location from the given 2D image data, which is reliable for this type of process. The brain tumor detection and segmentation is done with the threshold based anisotropic diffusion filter using bounding box method. This method achieves the result by comparing with the most recent related literatures. Overall, the process is done with the help of MATLAB with the adaptation of 2018a.