Lei Zhang








Level Set-based Image Segmentation with Global Shape Prior

Another early project I did in 2006 was developing a global shape prior for level-set based image segmentation. Different from the global shape prior defined with the region difference between two level set functions , we defined the global shape prior based on a contour difference, i.e., the difference between the the zero level-set of the evolving contour and the zero level-set of the global shape prior contour. We added such a global shape prior term into the "active contours without edges" (Chan & Vese 2001) image segmentation framework to form a segmentation approach. We applied this approach to segment both natural images and membrane images, and achieved good results. 


Example results:

Here are some typical results on segmenting membrane image sequences. 

Description: D:\Lei Zhang\MyWebpage\index_files\cryo(4000b)5.pngDescription: D:\Lei Zhang\MyWebpage\index_files\FIBtwomembranes113.png

Figure 1. Examples of membrane segmentation using the level-set based image segmentation with the proposed global shape prior. In the left image a single membrane was segmented, while in the right image double membranes were segmented.


Description: D:\Lei Zhang\MyWebpage\index_files\volume(4000)3.png

Description: D:\Lei Zhang\MyWebpage\index_files\volume(inner2000cv)2-shrink.PNG

Figure 2. The stacked membrane segmentation of multiple slices


Video Demos:

Here are some video clips of level set based image segmentation with global shape prior. (Note: If you cannot see these videos in your explorer, please try to use Windows Internet Explorer 6.0 or later version.)


Related publications:

§   Lei Zhang and Qiang Ji, "A Level Set-based Global Shape Prior and Its Application to Image Segmentation", in 4th IEEE International Workshop on Semantic Learning Applications in Multimedia, in conjunction with CVPR 2009 (oral presentation)

§   Lei Zhang and Qiang Ji, "A New Global Shape Prior for Level Set Based Segmentation", in 19th International Conference on Pattern Recognition (ICPR'08), 2008