Improved X-ray image segmentation based on GVF Snakes model
Xiang Binsen · Chinese Journal of Stereology and Image Analysis · 2011
At present,parametric active contour model has been more and more widely used in the segmentation of medical images,and gradient vector flow(GVF) snake model is one of the most widely studied algorithms.In the segmentation of the knuckles,the traditional GVF-Snake model needs some more control points,especially in the gap of the knuckles.In order to get more ideal curves,we propose a new algorithm based on the traditional GVF-Snake model.The improved algorithm firstly processes the image with a piecewise-linear method on the grey level of the image,and then redistributes the histogram to enhance the contrast of the image.The results prove that the new algorithm just needs several control points to get an accurate curve.So we can conclude that the improved algorithm is a better and more suitable algorithm for the segmentation of the knuckles.