Detection of an object within a volume of interest

Patent metadata

Field	Value
Publication number	US-10215717-B2
Application number	US-201514839883-A
Country	US
Kind code	B2
Filing date	Aug 28, 2015
Priority date	Aug 28, 2014
Publication date	Feb 26, 2019
Grant date	Feb 26, 2019

Abstract

Techniques, systems, and devices are disclosed for analyzing a point of closest approach (PoCA) image of a volume of interest (VOI) comprising a set of recorded PoCA points from charged particle detector measurements to detect an object within the VOI. The VOI is partitioned into a set of equally-sized bins with each bin including a subset of the PoCA points. A bin metric is determined for each bin. A subset of the bins is selected based on the detected bin metric with the subset of bins being most likely to contain objects. A potential object for each selected bin is determined by determining a location and a size for the potential object based at least on the PoCAs inside the bin. A figure of merit (FOM) of the potential object is determined as a measure of the likelihood that the potential object is truly a threat object.

First claim

What is claimed is:

1. A method for analyzing a point of closest approach (PoCA) image of a volume of interest (VOI) to detect one or more objects within the VOI, the method comprising:
detecting a plurality of charged particles using a charged particle detector and obtaining a set of recorded PoCA points based on the detected charged particles;
partitioning the VOI into a set of bins wherein each bin includes a subset of PoCA points of the set of recorded PoCA points;
calculating a bin metric for each bin in the set of bins, wherein the bin metric for each bin includes a median effective scattering angle;
selecting a subset of bins in the set of bins based on the calculated bin metric, wherein the subset of bins is most likely to contain objects; and
determining a potential object for each of the selected subset of bins by determining a location and a size for the potential object based at least on the PoCA points inside the bin.

2. The method of claim 1 , wherein the set of bins include equally-sized bins, and wherein the size of the bins is determined such that each of the set of bins contains on-average a same number of PoCA points, wherein the same number is a user-defined value.

3. The method of claim 1 , wherein the median effective scattering angle is calculated at least by:
calculating a product of a particle momentum and an associated scattering angle for each PoCA within the bin; and
calculating a median of the calculated products of the subset of PoCA points within each bin.

4. The method of claim 1 , wherein selecting the subset of bins in the set of bins based on the calculated bin metric includes selecting the top N bins in the set of bins with a highest product between scattered charged particles and the median effective scattering angle.

5. The method of claim 4 , wherein the subset of bins includes non-adjacent bins.

6. The method of claim 4 , wherein N is a user-selected number chosen to be large compared to the number of separate objects that are expected in the VOI, wherein the number of separate objects includes threatening objects, non-threatening objects of interest, and other non-threatening objects.

7. The method of claim 1 , wherein determining the location for a potential object includes determining a center of mass (CM) of each of the selected bin based on a momenta and scattering angles associated with the PoCA points inside the selected bin and PoCA points in the surrounding bins around the selected bin.

8. The method of claim 7 , wherein after determining the CM for a selected bin, the method further comprises:
re-centering the selected object to the location of the determined CM; and
re-calculating the bin metric for the selected object.

9. The method of claim 1 , further comprising:
filtering the set of recorded PoCA points to remove low value PoCA points corresponding to low scattering density prior to determining the size for a potential object.

10. The method of claim 9 , wherein filtering the set of recorded PoCA points to remove low value PoCA points includes determining a cutoff value for a product of a particle momentum and a scattering angle for a given PoCA point.

11. The method of claim 10 , wherein determining the cutoff value includes selecting PoCA points falling in a highest range of an expected distribution to achieve a user-specified constant false alarm rate.

12. The method of claim 10 , further comprising:
after filtering the set of recorded PoCA points, calculating the bin metric for a selected bin as the product of the median effective scattering angle and the number of charged particles in the selected bin, wherein the bin metric indicates the likely density and atomic number of the selected bin.

13. The method of claim 1 , wherein determining the size for a potential object includes determining a size separately for each of the x, y, and z directions, and for each direction:
constructing a histogram of the PoCA points inside the associated bin based on a product of a particle momentum and a scattering angle for each of the PoCA points inside the associated bin;
applying a smoothed density function on the histogram; and
measuring a distance between points on the smoothed density function that fall a predetermined value below the maximum value.

14. The method of claim 13 , wherein the predetermined value is substantially 3 dB.

15. The method of claim 1 , further comprising:
after determining the potential object for each of the selected subset of bins, determining a figure of merit (FOM) of the potential object as a measure of the likelihood that the potential object is truly a target object.

16. The method of claim 15 , wherein the FOM value is determined as the product of the median effective scattering angle times the total number of scattered charged particles within the potential object and divided by the product of the estimated volume of the potential object and the total number of particles transiting the volume.

17. The method of claim 16 , wherein the FOM value is an estimated density of scattering events times the median effective scattering angle of the events within the volume of the potential object.

18. The method of claim 1 , further comprising:
while calculating the bin metric for each bin in the set of bins, mitigating bin edge effect by performing the following:
shifting each bin by a portion of the linear dimension of the bin along each of the three axes x, y and z;
calculating the bin metric for each of the shifted bins and an unshifted bin; and
assigning a highest calculated bin metric among the shifted bins and the unshifted bin to the unshifted bin.

19. The method of claim 1 , further comprising correcting a shift of the center of mass location of a determined potential object from the location of the highest PoCA density.

20. The method of claim 19 , wherein correcting the shift of the center of mass location of the determined potential object includes:
obtaining a reconstructed charged particle image of the VOI, wherein the reconstructed charged particle image is partitioned into a set of voxels, and wherein each voxel is associated with a scattering density;
computing a modified center of mass location for the determined potential object based on both a first set of voxels occupying the volume of the determined potential object and a second set of voxels occupying a layer surrounding the determined potential object; and
moving the determined potential object to center on the modified center of mass location.

21. The method of claim 20 , wherein computing the modified center of mass location based on the first set of voxels and the second set of voxels involves computing a summation over the first and second sets of voxels of a product of the associated scattering density of each voxel and the vector location of the center of each voxel.

22. A system for analyzing a point of closest approach (PoCA) image of a volume of interest (VOI) to detect one or more objects within the VOI, the system comprising:
a processor;
a memory; and
an image processing mechanism communicatively coupled to the processor and the memory, wherein the image processing mechanism is configured to:
detect a plurality of charged particles using a charged particle detector and obtain a set of recorded PoCA points based on the detected charged particles;
partition the VOI into a set of bins wherein each bin includes a subset of PoCA points of the set of recorded PoCA points;
calculate a bin met

Assignees

• Decision Sciences Int Corp

Inventors

• Kurnadi Priscilla
• Mckenney Shawn
• Simon Sean
• Lam Peter

Classifications

• G01N23/046 Primary

  using tomography, e.g. computed tomography [CT] · CPC title

• G01V5/0016

  Physics · mapped topic

• G01N9/24

  by observing the transmission of wave or particle radiation through the material · CPC title

• G01V5/22

  Active interrogation, i.e. by irradiating objects or goods using external radiation sources, e.g. using gamma rays or cosmic rays · CPC title