A method of improved watershed segmentation algorithm
Peng Feng-ping · Journal of Guizhou Normal University · 2010
In order to extract brain tissue from human brain CT image,we propose an improved watershed algorithm.Firstly,we use K-means clustering to produce a primary segmentation,thus effectively inhibiting over-segmentation caused by grey level change of the surface of picture,which can be located at the contour more accurately.Then the improved watershed segmentation algorithm makes use of automated thresholding on the gradient magnitude map.Finally,regions merging algorithm is applied to avoid over-segmentation.The experimental results demonstrate the simplicity and effectivelness of the approach,in accordance with our understanding on the semantic concepts and the perceptual mechanism of human eyes.