—Positron emission tomography (PET) image segmentation is a difficult task due to low resolution, high contrast, large variability of pathology, high uncertainties in object boundaries and the inherent noise of PET images. A two staged automatic segmentation technique based on Gabor Annulus filtering and region growing is proposed here. Gabor Annulus filtering is applied on preprocessed image and the result is the input for region growing based on automatic seed selection and threshold determination. Unsupervised evaluation schemes are used for appraising the proposed technique. 93.5% of the lesions considered for validation have been obtained correctly in the segmented image.
—Gabor annulus filtering, intra-inter disparity, intra region uniformity, inter region contrast, positron emission tomography, region growing.
P. S. Arya Devi is with Department of Electronics, Model Engineering College, Kochi, Kerala, India (e-mail: email@example.com).
M. G. Mini is with College of Engineering, Cherthala, Kerala, India (email: firstname.lastname@example.org).
Cite:P. S. Arya Devi and M. G. Mini, "PET Image Segmentation Based on Gabor Annulus Filtering and Region Growing," International Journal of Computer Theory and Engineering vol. 8, no. 5, pp. 409-414, 2016.