The K-means Clustering Based Fuzzy Edge Detection Technique on MRI Images

Abstract

Edge detection plays a vital role in medical imaging applications such as MRI segmentation. Magnetic resonance imaging (MRI) is an imaging technique used in medical science to diagnose tumors of the brain by producing high quality images of the inside of the human body, by using various edge detectors. There exists many edge detector but still, need for research is felt in order to enhance their performance. A very common problem faced by most of the edge detector is the choice of threshold values. This paper presents fuzzy based edge detection using K-means clustering method. The K-means clustering approach is used in generating various groups which are then input to the mamdani fuzzy inference system. This whole process results in the generation of the threshold parameter which is then fed to the classical sobel edge detector which helps in enhancing its edge detection capability using the fuzzy logic. This whole setup is applied on the MR images of the human brain. The retrieved results represents that fuzzy based k-means clustering enhances the performance of classical sobel edge detector and along with retaining much relevant information about the tumors of the brain.