A Region Growing Segmentation Approach for MRI Brain Image Processing

Magnetic resonance imaging (MRI) is a crucial medical imaging methodology for diagnosing the disease. Image segmentation refers to the division of various sectors which is regarded as a significant job in medical image processing. This paper introduces an approach based on the region growing method for MRI brain image processing. In the pre-processing, the images will be filtered through anisotropic diffusion filtering algorithm so as to remove the noises and avoid the indistinctness. In the region growing stage, the threshold will be slightly increased and the objective function will be used for segmentation processes. The experimental results show the better performance of the proposed approach comparing with mean value filter and medium value filter methods.

[1]  Chang-Tsun Li,et al.  Weighted Level Set Evolution Based on Local Edge Features for Medical Image Segmentation , 2017, IEEE Transactions on Image Processing.

[2]  W. Marsden I and J , 2012 .

[3]  Daniel L. Rubin,et al.  Volumetric Image Registration From Invariant Keypoints , 2017, IEEE Transactions on Image Processing.

[4]  R. Nadarajan,et al.  CT and MRI image compression using wavelet-based contourlet transform and binary array technique , 2017, Journal of Real-Time Image Processing.