Abstract: In the field of medical image handling, location of brain tumor from Magnetic resonance image (MRI) examine has become to be an active amongst the most dynamic exploration. Detection of the tumor is the fundamental goal of the framework. In this paper, MRI brain image is utilized to tumor recognition process. This framework incorporates test the brain image process, image separating, segmentation, and morphological operation, tumor detection. In recent years many image processing procedures are widely used on medical images to detect tumors at an early and treatment stages. When the images are added with noise the difficulties are primarily associated with the detection of tumors. In such cases the intensity thresholding might fail to identify the tumors. Edges characterize boundaries are the fundamental importance in image processing. The image edge detection reduces the data by filtering and preserving the important structural attributes. So understanding the edge detection algorithms is essential. Here the morphology based region of interest segmentation combined with watershed transform of MRI tumor image is performed and comparative analysis in noisy environment such as Gaussian, salt and pepper, Poisson and speckle is performed. Here different edge detection filters such as average, Gaussian and laplacian, sobel and prewit, unsharp and log in presence of noise. The parameters such as Cross correlation, Peak signal to noise ratio, Mean absolute error for different operators are evaluated. The motivation behind the morphological operators is to separate the tumor part of the image. In this paper, the brain image is considered for examination and recognition. At first the region of interest is found, that recognizes the specific substance of the image and set the boundary of it. Morphological operations are utilized for edge identification. The four parameters, such as Probabilistic Rand Index (PRI), Variation of Information (VOI), Global Consistency Error (GCE), Structural similarity (SSIM) have been utilized to evaluate the performance of segmentation.

Keywords: Magnetic Resonance Image (MRI), Segmentation, Noise, Morphological Operation, Watershed Segmentation, Brain Tumor.