Anisotropic tubular minimal path model with fast marching front freezing scheme
作者：
Highlights：
• We establish a new anisotropic geodesic metric in a radius-lifted space in order to avoid the short branches combination problem.
• We extend the front frozen scheme from the isotropic metric to a radius-lifted space through an anisotropic Riemannian metric by using the non-local path feature.
• We make use of a crossing-adaptive tensor field established in the radius-lifted space, the anisotropy of which is kept in non-crossing vessel region and removed in crossing points.
• We consider three different types of scale function in the tensor field to reduce the influence from the regions outside the vessel structures including vesselness map, a binary-valued function and the skeleton map.
摘要
•We establish a new anisotropic geodesic metric in a radius-lifted space in order to avoid the short branches combination problem.•We extend the front frozen scheme from the isotropic metric to a radius-lifted space through an anisotropic Riemannian metric by using the non-local path feature.•We make use of a crossing-adaptive tensor field established in the radius-lifted space, the anisotropy of which is kept in non-crossing vessel region and removed in crossing points.•We consider three different types of scale function in the tensor field to reduce the influence from the regions outside the vessel structures including vesselness map, a binary-valued function and the skeleton map.
论文关键词：Minimal path model,Anisotropy enhancement,Riemannian metric,Path feature,Tubular structures
论文评审过程：Received 16 August 2018, Revised 10 January 2020, Accepted 25 March 2020, Available online 4 April 2020, Version of Record 18 April 2020.
论文官网地址：https://doi.org/10.1016/j.patcog.2020.107349
