Abstract—Segmentation is the process of extracting points, lines or regions, which are then used as inputs for complementary tasks such as registration, measurement, movement analysis, visualization , etc in MRI. The noise in MR images degrades the image quality and also affect on the segmentation process which can lead to wrong diagnosis. The main aim of this study is to suggest a system to enhance the quality of the human brain MRI. In the proposed system, median filter is used for image enhancement of brain MRI and fuzzy c-means for segmentation purpose. The proposed method is completely automatic that is there is no user involvement in the proposed system. The system is tested on different kinds of brain MR images and proved robust against noise as well as segments the images fast with improvements.
Index Terms—Dunn's index, fuzzy c-means (FCM), image segmentation, median filter.
Arshad Javed is with the Faculty of Computer Science and Information, Aljouf University Saudi Arabia and also with the Faculty of Computer Science and Information Technology, Universititi Malaysia Sarawak, Malaysia (e-mail: arsh_qau@ju.edu.sa).
Wang Yin Chai and Narayan Kulathuramaiyer are with the Faculty of Computer Science and Information Technology, Universititi Malaysia Sarawak, Malaysia (e-mail: ycwang@fit.unimas.my, nara@fit.unimas.my)
Abdulhameed Rakan Alenezi is with the Faculty of Computer Science and Information, Aljouf University Saudi Arabia (e-mail: a.alenezi@hotmail.com).
Cite:Arshad Javed, Wang Yin Chai, Abdulhameed Rakan Alenezi, and Narayan Kulathuramaiyer, "Enhancement of Magnetic Resonance Images Using Soft Computing Based Segmentation," International Journal of Machine Learning and Computing vol.4, no. 1, pp. 73-78, 2014.