Miao Yalin1, Zhang Hongmei1, Bian Zhengzhong1, Miao Xianglin2, Zhang Dalong1
(1. School of Life Science and Technology, Xi'an Jiaotong University, Xi'an 710049, China; 2. School of
Electronics and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China)
Abstract: A new approach named anisotropic reaction diffusion to vessels images
enhancement is proposed and then an efficient approach to thin vessels extraction based on
the curve evolution is adopted. The proposed approach facilitates enhancing the
directional structure of vessels and strengthening the edge while noise is removed in this
way. The curve evolution can not only drive the curve propagation in the low contrast and
thin vessel branches without the earlier boundary leaking, but also guarantee the curve
marching monotonically, and the curve evolution equation can be solved with fast marching
algorithm. Simultaneously, the multi-initial scheme is adopted for improving real time
implementation. The experiments show that thin pulmonary vessels at six to seven level can
be extracted fast and accurately.
Keywords: thin vessels; anisotropic; enhancement; deforming curve; fast marching;
extraction