A modified Kernelized Fuzzy C-Means (KFCM) algorithm for noisy images segmentation: Application to MRI images
Image segmentation is a low-level processing operation, it is the basis for many applications in both industrial vision, medical imaging. The approach provides a partition of the image by gathering pixels with similar gray levels in the same class of pixels. The main problem of this algorithm is that it does not take into account the topology of the image; it is based only on the value of pixels. Thus, it is very sensitive to noise and inhomogeneities in the image again, it remains dependent on the initialization of cluster centers.
In general the clustering algorithm chooses the initial centers in random manner but the initialization of cluster centers, using \Expectation Maximization\ algorithm allows an optimal choice of these centers. To account for the topology of the image, the statistical parameters of a window around the pixel are considered. These a priori are used in the cost function to optimize. The application of MRI images of the public database Brain Web, with different levels of noise, shows the performance of the proposed approach.
Keywords: image segmentation, FCM, KFCM, clustering, MRI images.
Download Full-Text
ABOUT THE AUTHORS
Abdenour Mekhmoukh
LTII Laboratory, Electrical Engineering department, University of Bejaia, Algeria
Karim Mokrani
LTII Laboratory, Electrical Engineering department, University of Bejaia, Algeria
Mohamed Cheriet
Department of Automated Manufacturing Engineering , ETS, Montreal, Quebec, Canada
Abdenour Mekhmoukh
LTII Laboratory, Electrical Engineering department, University of Bejaia, Algeria
Karim Mokrani
LTII Laboratory, Electrical Engineering department, University of Bejaia, Algeria
Mohamed Cheriet
Department of Automated Manufacturing Engineering , ETS, Montreal, Quebec, Canada
