Abstract
AN IMPROVED WATERSHED ALGORITHM FOR MEDICAL IMAGE SEGMENTATION
*Dr. Ajala Funmilola A., Falohun A. S. and Opasola T. O.
ABSTRACT
This paper improves watershed segmentation algorithm to overcome its over-segmentation challenges using wavelet transform method at its post- possessing stage. The existing and improved algorithms (watershed algorithm and the improved watershed algorithm) were tested on medical images and comparative analysis in terms of segmentation accuracy, segmentation time and memory consumption was considered. Thirty (30) medical images were used for analysis and the average percentage of segmentation accuracy, segmentation time and memory consumption for existing watershed are 35%, 53% and 53%, respectively. The average percentage of segmentation accuracy, segmentation time and memory consumption of improved watershed for thirty (30) medical images are 65%, 47% and 47%, respectively. Based on the results obtained, the accuracy of the improved watershed was 30% higher than existing watershed while the segmentation time and memory consumption of the existing watershed is lower than the improved watershed by 6% and 6%, respectively.
[Full Text Article] [Download Certificate]