Multidimensional segmentation based on adaptive bounding box and ellipsoid models
View Patent ↗A seed point is selected inside a structure that is to be segmented in image data. An adaptive model is defined around the seed point, and a preprocessing filter is applied only within the bounding region. A presegmentation of the preprocessed result is performed, and the bounding region is expanded if necessary to accommodate the presegmentation result. An adaptive model for post-processing may be used. The model is translated, rotated and scaled to find a best fit with the pre-segmented data.
1. A method for segmenting a structure in a set of image data, the structure being identified by one or more seed points, comprising a processor configured to perform the following steps of:
a) defining an adaptive model around the seed point having one or more boundaries and sectors;
b) applying prefiltering only on the area defined by the adaptive model;
c) applying presegmentation within the result of prefiltering; and,
d) analyzing results of presegmentation for rotation or translation of the adaptive model;
e) if the result of the presegmentation touches any of the one or more boundaries of the adaptive model, expanding the corresponding boundary and sector.
2. The method as claimed in claim 1 , further-comprising repeating steps (b) to (e) until the region does not touch any of the one or more boundaries and generating a final segmentation of the structure.
3. The method as claimed in claim 2 , wherein the adaptive model is a bounding box.
4. The method as claimed in claim 2 , wherein the adaptive model is elliptic.
5. A method for segmenting a structure in a set of image data, the structure being identified by one or more seed points, comprising a processor configured to perform the following steps of:
(a) presegmenting the structure;
(b) defining a bounding shape around the seed point, the bounding shape having one or more boundaries;
(c) filtering at least a portion of the set of image data;
(d) determining whether any part of the one or more boundaries are within the bounding shape; and
(e) if any part of the one or more boundaries are within the bounding shape, growing a region within the bounding shape and if the region touches any of the one or more boundaries before reaching the structure, expanding the bounding shape.
6. The method as claimed in claim 5 , further comprising repeating steps (d) and (e) until the region does not touch any of the one or more boundaries and generating a final segmentation of the structure.
7. The method as claimed in claim 5 , wherein the presegmentation step is performed using one of the following segmentation procedures: region growing, graph cuts, or snake.
8. The method as claimed in claim 6 , further comprising outputting the final segmentation to an output device.
9. The method as claimed in claim 6 , wherein the bounding shape is a box.
10. The method as claimed in claim 6 , wherein the bounding shape is elliptic.
11. The method as claimed in claim 6 , wherein the filtering step is performed only on the set of image data located within the bounding shape with an edge preserving smoothing filter.
12. The method as claimed in claim 11 , wherein an anisotropic diffusion filter is used to filter the image data.
13. A computer system for segmenting a structure in a set of image data, the structure being identified by one or more seed points, comprising:
one or more processors;
a memory readable by the processor, the memory comprising program code executable by the processor, the program code adapted to perform the following steps:
(a) presegmenting the structure;
(b) defining a bounding shape around the seed point, the bounding shape having one or more boundaries;
(c) filtering at least a portion of the set of image data;
(d) determining whether any part of the one or more boundaries are within the bounding shape; and
(e) if any part of the one or more boundaries are within the bounding shape, growing a region within the bounding shape and if the region touches any of the one or more boundaries before reaching the structure, expanding the bounding shape.
14. The system as claimed in claim 13 , wherein the program code is further adapted to repeat steps (d) and (e) until the region does not touch any of the one or more boundaries and generating a final segmentation of the structure.
15. The system as claimed in claim 13 , wherein the presegmentation is performed using one of the following segmentation procedures: region growing, graph cuts, or snake.
16. The system as claimed in claim 14 , further comprising a display that displays the final segmentation.
17. The system as claimed in claim 14 , wherein the bounding shape is a box.
18. The system as claimed in claim 14 , wherein the bounding shape is elliptic.
19. The system as claimed in claim 14 , wherein the filtering is performed only on the set of image data located within the bounding shape with an edge preserving smoothing filter.
20. The system as claimed in claim 19 , wherein an anisotropic diffusion filter is used to filter the image data.