Object detection using cross-section analysis

1. A system for 3-D object detection comprising:

a memory device for storing a program; and

a processor in communication with the memory device, the processor operative with the program to perform a method, the method comprising:

pre-processing an image;

using cross-section analysis, building a confidence array; and

detecting a plurality of peaks in the confidence array, wherein peaks signify a likelihood of a 3-D object of interest,

wherein cross-section analysis comprises:

obtaining a plurality of cross sections of an image;

defining a confidence array; and

analyzing the cross sections of the image to identify 2-D objects of interest, accumulating evidence in the confidence array, and

wherein cross-section analysis is performed on cross sections perpendicular to the x-, y-, and z-dimensions, respectively.

2. The system of claim 1 , wherein pre-processing an image comprises thresholding the image.

3. The system of claim 1 , wherein cross-section analysis is performed on cross sections evenly distributed among different directions.

4. The system of claim 1 ,wherein analyzing the cross sections of the image comprises analyzing 2-D shapes in 2-D cross section images.

5. The system of claim 4 , wherein analyzing 2-D shapes in 2-D cross section images comprises using boundary analysis.

6. The system of claim 1 , wherein detecting peaks in the confidence array comprises:

applying a low-pass filter to the confidence array; and

selecting peaks.

7. The system method of claim 1 , wherein the object is a compact round-shaped object.

8. A cross-section analysis system for nodule candidate generation comprising:

a memory device for storing a program: and

a processor in communication with the memory device, the processor operative with the program to perform a method, the method comprising:

pre-processing an image;

analyzing 2-D object boundaries, obtaining convex boundary segments:

classifying a plurality of points on each convex boundary segment as 2-D peak points or 2-D convex points;

classifying surface points as 3-D peak points, 3-b convex points or normal points;

establishing a 3-D confidence array; and

detecting a plurality of peaks in the confidence array, wherein peaks signify a likelihood of a 3-D object of interest.

9. The cross-section analysis system of claim 8 , wherein pre-processing an image comprises thresholding the image.

10. The cross-section analysis system of claim 8 , wherein analyzing 2-D object boundaries comprises:

tracing a boundary of a foreground object in a cross section;

breaking the boundary into a number of curve segments using curvature; and

identifying convex boundary segments.

11. The cross-section analysis system of claim 10 , wherein each of the convex boundary segments have a normalized curvature larger than a predetermined value.

12. The cross-section analysis system of claim 8 , wherein a point is classified as a 3-D peak point if the point is a 2-D peak point in all its 2-D cross sections.

13. The cross-section analysis system of claim 8 , wherein a point is a 3-D convex point if (a) it is not a 3-D peak point and (b) it is a 2-D peak point or 2-D convex point in all its 2-D cross sections.

14. A computer readable medium including computer code for 3-D object detection, the computer readable medium comprising:

computer code for preprocessing an image;

computer code for using cross-section analysis to build a confidence array; and

computer code for detecting a plurality of peaks in the confidence array, wherein peaks signify a likelihood of a 3-D object of interest,

wherein the computer code for using cross-section analysis to build a confidence array comprises:

computer code for obtaining a plurality of cross-sections of images;

computer code for defining a confidence array; and

computer code for performing analysis of the cross-section images to identify 2-D objects of interest to accumulate evidence in the confidence array and

wherein cross-section analysis is performed on cross sections perpendicular to the x-, y-, and z-dimensions, respectively.

15. A computer readable medium including computer code for nodule candidate generation, the computer readable medium comprising:

computer code for preprocessing an image;

computer code for analyzing 2-D object boundaries to obtain convex boundary segments;

computer code for classifying a plurality of points on each convex boundary segment as 2-D peak points or 2-D convex points;

computer code for classifying surface points as 3D peak points, 3-D convex points or normal points;

computer code for establishing a 3D confidence array; and

computer code for detecting a plurality of peaks in the confidence array, wherein peaks signify a likelihood of a 3-D object of interest.