| 摘要 |
A technique is described for enhancing three dimensional or volumetric image data, such as thick slice or slab data, or data for slices stacked in a third dimension orthogonal to an imaging plane. The technique processes the image data by reference to data parameters both in two dimensions and in three dimensions. The processing permits identification of structural pixels and their differentiation from non-structural pixels. The structural pixels may be identified by reference to gradients determined in three dimensions, but with directions determined by reference to only two dimensions. The structural and non-structural pixels may then be processed differently to provide an enhanced image.
|
