| Very large 3D digital images of arterial trees can be produced by many imaging scanners. While many automatic approaches have been proposed that can begin the process of defining the 3D arterial tree captured in such an image, none guarantee complete, accurate definition. This thesis proposes automatic, semi-automatic and 3D graphical interactive techniques for coming closer to the ultimate goal of defining a complete and accurate 3D arterial tree.; From our point of view, the primary drawback of previously proposed skeletonization and central-axis methods are the modeling of axes in junction areas of branching structures and the excessive number of false branches. One of our goals is a fully automatic central-axes algorithm capable of dealing with the variety of structures depicted in an image of a 3D tree. In addition, we focus on developing interactive editing tools in a 3D visualization system, integrated with the automatic approaches. We design a semi-automatic skeleton analyzer to help identify and repair errors. The results lead to the requirements for completing quantitative methods. The integrated effort reaches the goal of accurate multi-generational tree analysis.; We provide numerous results from 3D micro-CT imaging and 3D lung imaging. The results show the improvement offered by our methods over previous methods. |
