Method and apparatus for enhancing an image using data optimization and segmentation

A method of grouping image data points based on smoothness or roughness values, the method comprising:

• creating a reduced-dimensionality image data set from a multidimensional image data set;
• selecting a first set of points in the reduced-dimensionality image data set, each point in the first set of points having a corresponding point in a second set of points in the multidimensional image data set;
• defining a unique projection path for each point in the first set of points, the projection path extending in a direction Z from a point in the first set of points through the corresponding point in the second set of points;
• determining a distance measure for a first point in the first set of points, the distance measure being the distance along the projection path in the Z direction between the first point in the first set of points and the first point's corresponding point in the second set of points;
• determining the distance measures for multiple points, including an image point, in the first set of points;
• calculating a smoothness or roughness value for the image point in the first set of points by comparing the distance measure of the image point to the distance measures of other points in the first set of points;
• and
• grouping the image point with similar points in the first set of points, each of said similar points having both a smoothness or roughness value related to the smoothness or roughness value of the image point.