| Abstract: | This paper introduces a novel segmentation scheme
based on multi-resolution analysis and watershed
segmentation algorithm. The proposed scheme consists
of six major stages. These stages are 1) pre-processing, 2)
wavelet transform, 3) image denoising, 4) watershed
segmentation, 5) segmented image projection and finally
6) boundary extraction. Experimental results
demonstrated the efficiency of the proposed scheme in
segmenting carotid artery ultrasound images, where the
computational cost of the watershed-based segmentation
scheme is reduced, as it is applied to a small
low-resolution image. At the same time, the segmentation
accuracy is increased as the proposed scheme is more
robust to noise and hence, it prevents over segmentation
in final segmented images.
|