USING MATHEMATICAL MORPHOLOGY FOR CORRECTION OF MAGNETIC RESONANCE IMAGES

SUMMARY The paper presents a morphological method for brightness correction of Magnetic Resonance (MR) images, which makes possible the use of watershed technique for image segmentation and extraction of objects. As an example of image correction, extraction of the mask of the gray matter from the image of the human spinal cord is given. The described image correction is based on the use of the White Top Hat (WTH) transform with a large structuring element. As a result of the correction, brightness in the image becomes more homogeneous, and one can merge the watershed regions included in the mask of the gray matter in a semi-automatic procedure. This is a significant improvement in comparison with uncorrected images, for which development of such a procedure proved to be very difficult. The current paper is a simplified and abbreviated version of [10].

[1]  Mariusz Nieniewski,et al.  Vision system identification of large defects on the surface of ferrite cores , 2000 .

[2]  Luc Vincent,et al.  Watersheds in Digital Spaces: An Efficient Algorithm Based on Immersion Simulations , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Hans Burkhardt,et al.  A Parallel Watershed Algorithm , 1996 .

[4]  Bidyut Baran Chaudhuri,et al.  An efficient method based on watershed and rule-based merging for segmentation of 3-D histo-pathological images , 2001, Pattern Recognit..

[5]  Pierre Soille,et al.  Morphological Image Analysis , 1999 .

[6]  Alan C. Evans,et al.  A nonparametric method for automatic correction of intensity nonuniformity in MRI data , 1998, IEEE Transactions on Medical Imaging.

[7]  Richard A. Robb,et al.  Optimized homomorphic unsharp masking for MR grayscale inhomogeneity correction , 1998, IEEE Transactions on Medical Imaging.

[8]  Sinan Batman,et al.  Morphological Methods for Biomedical Image Analysis , 2000 .

[9]  Luc Vincent,et al.  Determining watersheds in digital pictures via flooding simulations , 1990, Other Conferences.

[10]  Mariusz Nieniewski,et al.  Segmentation of spinal cord images by means of watershed and region merging together with inhomogeneity correction , 2002 .

[11]  F. Zana,et al.  A region merging algorithm using mathematical morphology: application to macula detection , 1998 .

[12]  Benoit M. Dawant,et al.  Correction of intensity variations in MR images for computer-aided tissue classification , 1993, IEEE Trans. Medical Imaging.

[13]  Reinhard Klette,et al.  Handbook of image processing operators , 1996 .

[14]  Roger Serneels,et al.  On a hybrid watershed-neural network approach for the segmentation of high field MR images of the spinal cord , 1997 .

[15]  M. Stella Atkins,et al.  Segmentation of multiple sclerosis lesions in intensity corrected multispectral MRI , 1996, IEEE Trans. Medical Imaging.

[16]  Etienne Kerre,et al.  Segmentation of images using a simplified watershed algorithm , 2000 .

[17]  Luc Vincent,et al.  Morphological grayscale reconstruction in image analysis: applications and efficient algorithms , 1993, IEEE Trans. Image Process..