Multilevel image segmentation
First Claim
1. A method for processing a full resolution CT scan image comprising:
- a) generating at a computer a reduced resolution image from said full resolution CT scan image;
b) segmenting at said computer said reduced resolution image to identify a boundary between an object and a background;
c) generating at said computer an increased resolution boundary image, said increased resolution boundary image corresponding to only a portion of an area of said full resolution CT scan image, based on said prior identified boundary; and
d) segmenting at said computer said increased resolution boundary image to generate a refined boundary by;
generating a graph of said increased resolution boundary image, wherein said step of generating a graph for said increased resolution boundary image uses an outer layer of said increased resolution boundary image as object seeds and the inner layer of said increased resolution boundary image as background seeds, andperforming a graph cut on said graph.
5 Assignments
0 Petitions
Accused Products
Abstract
A multilevel image segmentation technique using graph cuts is disclosed. A reduced resolution image is generated from a full resolution image which is to be segmented. The reduced resolution image is then segmented in order to identify a boundary between an object and a background within the image. The identified boundary then identifies a portion of an increased resolution image which is segmented in order to refine the earlier identified boundary. The steps may be iterated for successively increasing image resolutions in order to refine the boundary as required by a particular application. An initial identification of object and background portions of the image may be provided as input by a user. Alternatively, a user may identify only the object portion, and the background portion may be automatically determined.
33 Citations
25 Claims
-
1. A method for processing a full resolution CT scan image comprising:
-
a) generating at a computer a reduced resolution image from said full resolution CT scan image; b) segmenting at said computer said reduced resolution image to identify a boundary between an object and a background; c) generating at said computer an increased resolution boundary image, said increased resolution boundary image corresponding to only a portion of an area of said full resolution CT scan image, based on said prior identified boundary; and d) segmenting at said computer said increased resolution boundary image to generate a refined boundary by; generating a graph of said increased resolution boundary image, wherein said step of generating a graph for said increased resolution boundary image uses an outer layer of said increased resolution boundary image as object seeds and the inner layer of said increased resolution boundary image as background seeds, and performing a graph cut on said graph. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. An apparatus for processing a full resolution image comprising:
-
a processor; and a non-transitory computer readable medium storing computer program instructions, which when executed by the processor cause the processor to perform operations comprising; a) generating a reduced resolution image from said full resolution image; b) segmenting said reduced resolution image to identify a boundary between an object and a background; c) generating an increased resolution boundary image, said increased resolution boundary image corresponding to only a portion of an area of said full resolution image, based on said prior identified boundary; and d) segmenting said increased resolution boundary image to generate a refined boundary by; generating a graph of said increased resolution boundary image, wherein said generating a graph for said increased resolution boundary image uses an outer layer of said increased resolution boundary image as object seeds and the inner layer of said increased resolution boundary image as background seeds, and performing a graph cut on said graph. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A method for segmenting a CT scan image (I0) based on object seeds O and background seeds G comprising the steps of:
-
a) generating at a computer a plurality (K) of reduced resolution images (I1, . . . , IK); b) generating at said computer a graph GK for the lowest resolution image IK; c) calculating at said computer a minimum cut CK of said graph GK based on O and G; d) generating at said computer a binary boundary image Jk to represent the image points identified by nodes in said minimum cut CK; e) projecting at said computer said image points onto a higher resolution boundary image Jk−
1;f) generating at said computer a graph GK−
1 for said higher resolution boundary image Jk−
1 ; andg) calculating at said computer a minimum cut CK−
1 for said graph GK−
1,wherein said step of generating a graph GK−
1 for said higher resolution boundary image Jk−
1 uses the outer layer of said higher resolution boundary image as object seeds (O) and the inner layer of said higher resolution boundary image as background seeds (G). - View Dependent Claims (22, 23, 24, 25)
-
Specification