Image based object locator

Patent metadata

Field	Value
Publication number	US-10115199-B2
Application number	US-201514876380-A
Country	US
Kind code	B2
Filing date	Oct 6, 2015
Priority date	Oct 8, 2014
Publication date	Oct 30, 2018
Grant date	Oct 30, 2018

Abstract

Techniques, systems, and devices are disclosed for analyzing a reconstructed charged particle image of a volume of interest from charged particle detector measurements to determine a location and boundaries of one or more objects or an orientation of the one or more objects. The technique can include performing a segmentation operation on the reconstructed charged particle image of the volume. The segmentation operation identifies a subset of a set of voxels of the image of the volume as object candidate voxels. The technique can include locating corners of the one or more objects to determine the location, boundaries, or the orientation of the one or more objects. The technique can also include the computation of the center of mass of the one or more objects. The technique can include performing a morphological operation on the image and can include performing a connected-component analysis on the identified object-candidate voxels.

First claim

What is claimed is:

1. A method for analyzing a reconstructed charged particle image of a volume from charged particle detector measurements to determine a location and boundaries of at least one object associated with the volume, the method comprising:
performing a segmentation operation on the reconstructed charged particle image of the volume, wherein the segmentation operation identifies a first set of voxels of the reconstructed charged particle image of the volume as object candidate voxels and identifies a second set of voxels as non-object voxels, wherein
the first set of voxels have intensity values that fall between a high threshold and a low threshold, and
the second set of voxels have intensity values above the high threshold or below the low threshold;

identifying additional object candidate voxels from the second set of voxels to fill holes of the first set of voxels;
locating edges of the object to allow determining an extent and an orientation of the object; and
based on location of the edges, determining the location, boundaries, and corners of the object.

2. The method of claim 1 , wherein the object includes at least one container; and
wherein determining the location, boundaries and corners of the object includes determining location, boundaries and corners of the at least one container in addition to location, boundaries and corners of contents of the at least one container.

3. The method of claim 1 , wherein the object includes at least one package; and
wherein determining the location, boundaries and corners of the object includes determining location boundaries and corners of contents of the at least one package.

4. The method of claim 1 wherein identifying additional object candidate voxels comprises:
performing a connected-component analysis on the identified object-candidate voxels.

5. The method of claim 4 further comprising:
prior to performing the connected-component analysis on the identified object-candidate voxels, performing a morphological operation on the constructed charged particle image.

6. The method of claim 5 , wherein performing the morphological operation on the reconstructed charged particle image involves applying a sequence of morphological dilation operations and morphological erosion operations to the at least one of the object-candidate voxels.

7. The method of claim 1 further comprising:
determining a relative position of some parts of the object relative to other parts of the object.

8. The method of claim 1 further comprising:
determining locations of the previously segmented fiducials by computing the center of mass of each fiducial and computing the center of mass of the one or more objects.

9. The method of claim 1 , wherein the reconstructed charged particle image includes one of:
a cosmic-ray muon image;
a cosmic-ray electron image; or
a combined cosmic-ray muon and cosmic-ray electron image.

10. The method of claim 1 , comprising:
determining the orientation of the object based on the location of the edges.

11. A system for analyzing a reconstructed charged particle image of a volume from charged particle detector measurements to determine a location and boundaries of at least one object associated the volume, the system comprising:
a processor;
a memory; and
an image processing mechanism coupled to the processor and the memory, wherein the image processing mechanism is configured to:
perform a segmentation operation on the reconstructed charged particle image of the volume, wherein the segmentation operation identifies a first set of voxels of the reconstructed charged particle image of the volume as object candidate voxels and identifies a second set of voxels as non-object voxels, wherein
the first set of voxels have intensity values that fall between a high threshold and a low threshold, and
the second set of voxels have intensity values above the high threshold or below the low threshold;

identify additional object candidate voxels from the second set of voxels to fill holes of the first set of voxels;
locate edges of the object to allow a determination of an extent and an orientation of the object and
based on location of the edges, determine the location, boundaries and corners of the object.

12. The system of claim 11 , wherein the image processing mechanism is configured to determine the orientation of the object in addition to the location, boundaries and corners, based on the location of the edges.

13. The system of claim 11 , wherein the object includes at least one container and determining the location, boundaries and corners of the object includes determining a location, boundaries and corners of the at least one container in addition to a location, boundaries and corners of contents of the container.

14. The system of claim 11 , wherein the object includes at least one package and determining the location and boundaries of the object includes determining boundaries of contents of the at least one package.

15. The method of claim 1 , further comprising determining an average intensity of the connected voxels within the object and a variance of voxel intensity that describes how much voxel intensities vary within the object.

16. The system of claim 11 , wherein the image processing mechanism is further configured to determine an average intensity of the connected voxels within the object and a variance of voxel intensity that describes how much voxel intensities vary within the object.

17. The system of claim 11 , wherein the image processing mechanism is further configured to:
perform a morphological operation on at least one of the object candidate voxels; and
perform a connected-component analysis on at least one of the object candidate voxels after the morphological operation.

Assignees

• Decision Sciences Int Corp

Inventors

• Patnaik Rohit

Classifications

• G06T12/30

  Image post-processing, e.g. metal artefact correction · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06K9/2054

  Physics · mapped topic

• G01V5/0016

  Physics · mapped topic

• G06T7/12

  Edge-based segmentation · CPC title