We propose a local feature called three dimensional (3D) concentration index for analyzing line texture patterns in 3D space. This feature can be used for evaluating quantitatively local concentration of lines at each point of a 3D line pattern. In the paper, simulations using artificial patterns are performed to examine basic properties of the index. The index is also applied to patterns of vessels and bronchi extracted from practical chest X-ray CT images. As a result, it is shown that the 3D concentration index can be used to quantify the degree of convergence of vessels and bronchi to a tumor
Published in:
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
(Volume:2
)
Date of Conference: 16-20 Aug 1998