Morphological edge detection and brain tumor segmentation in Magnetic Resonance (MR) images based on region growing and performance evaluation of modified Fuzzy C-Means (FCM) algorithm

Abstract

The medical image processing has become indispensable with an increased demand for systematic and efficient detection of brain tumor in a short period of time. There are various techniques for medical image segmentation. Detecting a wide variety of brain images in terms of shape and intensity is a challenging and difficult task to bring out a reliable and authentic data for diagnosing brain tumor diseases. This paper presents an algorithm which combines Region of Interest (ROI), Region Growing and Morphological Operation (Dilation and Erosion). This method initially identifies the approximate Region Growing (RG). Region growing is a procedure that groups pixels into larger regions, which starts from the seed points. Region growing based techniques are better than the edge-based techniques in noisy images where edges are difficult to detect. The Morphological Edge Detection of the input image is done and the input image is reconstructed on the basis of dilation and erosion for the enhancement of the image. The proposed work is divided into preprocessing to reduce the noise, Fuzzy C-Means is used to Region growing, Morphological edge detection is to enhance the image. Then the morphological edge detection can be classified into two categories, one is dilation and another is Erosion. Finally apply Gaussian filter to get output. After that, Fuzzy C-Means clustering (FCM), followed by seeded region growing is applied to detect and segment the tumor from the brain MRI image.