System and method for detecting objects with differential operators
View Patent ↗A system and a method for detecting an object, such as an explosive device or material, located within a closed article, such as a piece of luggage or a parcel. The system includes an acquisition subsystem for acquiring information pertaining to a specific object, a reconstruction subsystem for reconstructing acquired information pertaining to the specific object into image data, and a computer-aided detection subsystem adapted for identifying the specific object through the use of differential operators. The method includes obtaining image data of the one object, computing a differential operator for each voxel of the image data, computing eigenvalues and eigenvectors for each of the voxels, and computing a scalar function of the eigenvalues to ascertain whether each of the voxels represents a portion of the one object.
1. A system for distinguishing a specific object from other objects residing within an enclosed container, comprising:
an acquisition subsystem for acquiring information pertaining to a specific object;
a reconstruction subsystem for reconstructing acquired information pertaining to the specific object into image data; and
a computer-aided detection subsystem adapted for identifying the specific object through differential operators.
2. The system of claim 1 , wherein the acquisition subsystem comprises a computed tomography machine.
3. The system of claim 2 , wherein the computed tomography machine comprises one from the group consisting of a 3 rd generation computed tomography scanner, a 4 th generation computed tomography scanner, a 5 th generation computed tomography scanner, and a computed tomography scanner including a stationary x-ray source and a stationary detector.
4. The system of claim 1 , wherein the computer-aided detection subsystem is configured to identify the specific object as being one from the group consisting of locally sheet-like, spherical, cylindrical, and blob-like.
5. The system of claim 1 , wherein the differential operators comprise a Hessian tensor.
6. The system of claim 1 , wherein the differential operators comprise a Curvature tensor.
7. The system of claim 1 , wherein the computer-aided detection subsystem comprises more than one stage for distinguishing the specific object from the other objects.
8. The system of claim 7 , wherein at least one of the more than one stage computes a differential operator for each voxel of the image data.
9. The system of claim 8 , wherein at least one of the more than one stage computes eigenvalues and eigenvectors of the differential operators.
10. The system of claim 9 , wherein at least one of the more than one stage computes a scalar function of the eigenvalues to ascertain whether each of the voxels of the image data represents a portion of the specific object.
11. The system of claim 10 , wherein the scalar function responds to a greater degree to locally sheet-like materials than to materials having other geometries.
12. The system of claim 11 , wherein the scalar function is equal to the negative of the largest magnitude eigenvalue minus the product of a user selectable constant times the addition of two additional eigenvalues.
13. A system for distinguishing a specific object from other objects residing within an enclosed container, comprising:
an acquisition subsystem for acquiring information pertaining to a specific object, wherein the acquisition subsystem comprises a computed tomography machine;
a reconstruction subsystem for reconstructing acquired information pertaining to the specific object into image data; and
a computer-aided detection subsystem adapted for identifying the specific object through differential operators, wherein the computer-aided detection subsystem comprises more than one stage.
14. The system of claim 13 , wherein the computed tomography machine comprises one from the group consisting of a 3 rd generation computed tomography scanner, a 4 th generation computed tomography scanner, a 5 th generation computed tomography scanner, and a computed tomography scanner including a stationary x-ray source and a stationary detector.
15. The system of claim 13 , wherein the computer-aided detection subsystem is configured to identify the specific object as being one from the group consisting of locally sheet-like, spherical, cylindrical, and blob-like.
16. The system of claim 13 , wherein the differential operators comprise a Hessian tensor.
17. The system of claim 13 , wherein the differential operators comprise a Curvature tensor.
18. The system of claim 13 , wherein at least one of the more than one stage computes a differential operator for each voxel of the image data.
19. The system of claim 18 , wherein at least one of the more than one stage computes eigenvalues and eigenvectors of the differential operators.
20. The system of claim 19 , wherein at least one of the more than one stage computes a scalar function of the eigenvalues to ascertain whether each of the voxels of the image data represents a portion of the specific object.
21. The system of claim 20 , wherein the scalar function responds to a greater degree to locally sheet-like materials than to materials having other geometries.
22. The system of claim 21 , wherein the scalar function is equal to the negative of the largest magnitude eigenvalue minus the product of a user selectable constant times the addition the two additional eigenvalues.
23. A system for identifying a specific object from other objects resident within an enclosed container, comprising:
an acquisition subsystem for acquiring information pertaining to a specific object;
a reconstruction subsystem for reconstructing acquired information pertaining to the specific object into image data; and
a means for identifying the specific object through the use of differential operators, wherein vectors derived from the differential operators are used in the identification of the specific object.
24. The system of claim 23 , wherein the acquisition subsystem comprises a computed tomography machine from the group consisting of a 3 rd generation computed tomography scanner, a 4 th generation computed tomography scanner, a 5 th generation computed tomography scanner, and a computed tomography scanner including a stationary x-ray source and a stationary detector.
25. The system of claim 23 , wherein the means for identifying the specific object through the use of differential operators comprises a computer-aided detection subsystem configured to identify the specific object as being one from the group consisting of locally sheet-like, spherical, cylindrical, and blob-like.
26. The system of claim 25 , wherein the computer-aided detection subsystem comprises more than one stage for distinguishing the specific object from the other objects.
27. The system of claim 26 , wherein at least one of the more than one stage computes eigenvalues and eigenvectors of the differential operators.
28. The system of claim 27 , wherein at least one of the more than one stage computes a scalar function of the eigenvalues to ascertain whether each of the voxels of the image data represents a portion of the specific object.
29. The system of claim 23 , wherein the differential operators comprise a Hessian tensor.
30. The system of claim 23 , wherein the differential operators comprise a Curvature tensor.
31. A method for distinguishing a specific object from other objects within an enclosed container, comprising the steps of:
obtaining image data of a specific object;
computing a differential operator for each voxel of the image data;
computing eigenvalues and eigenvectors for each of the voxels; and
computing a scalar function of the eigenvalues to ascertain whether each of the voxels represents a portion of the specific object.
32. The method of claim 31 , wherein said obtaining step comprises:
acquiring information pertaining to the specific object; and
reconstructing the acquired information pertaining to the specific object into image data.
33. The method of claim 32 , wherein said acquiring step comprises acquiring information through a computed tomography machine.
34. The method of claim 31 , wherein the computing a differential operator step comprises computing a differential operator for each voxel of the image data with at least one stage of a computer-aided detection subsystem.
35. The method of claim 34 , wherein the computing eigenvalues and eigenvectors step comprises computing eigenvalues and eigenvectors of the differential operators with the at least one stage.
36. The method of claim 31 , wherein the computing a differential operator step comprises computing a Hessian tensor.
37. The method of claim 31 , wherein the computing a differential operator step comprises computing a Curvature tensor.
38. The method of claim 31 , wherein the scalar function responds to a greater degree to locally sheet-like materials than to materials having other geometries.
39. The method of claim 31 , further comprising identifying the one object as being from the group of objects consisting of locally sheet-like, spherical, cylindrical, and blob-like objects.
40. A method for distinguishing one locally sheet-like object from other objects within an enclosed container, comprising the steps of:
obtaining image data of a locally sheet-like object;
computing a differential operator for each voxel of the image data with at least one stage of a computer-aided detection subsystem;
computing eigenvalues and eigenvectors for each of the voxels; and
computing a scalar function of the eigenvalues to ascertain whether each of the voxels represents a portion of the locally sheet-like object;
wherein the scalar function responds to a greater degree to locally sheet-like materials than to materials having other geometries.
41. The method of claim 40 , wherein said obtaining step comprises:
acquiring information pertaining to the one object; and
reconstructing the acquired information pertaining to the one object into image data.
42. The method of claim 41 , wherein said acquiring step comprises acquiring information through a computed tomography machine.
43. The method of claim 40 , wherein the computing eigenvalues and eigenvectors step comprises computing eigenvalues and eigenvectors of each differential operator with the at least one stage.
44. The method of claim 40 , wherein the computing a differential operator step comprises computing a Hessian tensor.
45. The method of claim 40 , wherein the computing a differential operator step comprises computing a Curvature tensor.