Screening technique for prohibited objects at security checkpoints

1. A method for automatically classifying compartments at a security checkpoint, comprising:

receiving, by a processor of a computing system, a three-dimensional representation of a compartment from an imaging device connected to the computing system; and

classifying, by the processor, the compartment into a first category or a second category using a trained neural network that analyzes the three-dimensional representation, regardless of a shape or a material of an object contained within the compartment, without requiring human inspection of an output of the imaging device, wherein compartments classified into the first category are automatically cleared at the security checkpoint and compartments classified into the second category undergo further security checks.

2. The method of claim 1 , wherein the three-dimensional representation is a set of images in sequence of the interior region of the compartment, output by a computed tomography (CT) scanner, further wherein each image in the set of images has a width and a height, and when stacked together, form a depth.

3. The method of claim 1 , wherein the trained neural network is trained by:

calculating, by the processor, a maximum intensity projection across each dimension of at least two orthogonal projections of a known three-dimensional representation of a prohibited object;

stacking, by the processor, each of the maximum intensity projections into a single image; and

feeding, by the processor, the single image into the trained neural network so that the trained neural network learns that the single image having the same maximum intensity projections corresponds to the prohibited object.

4. The method of claim 3 , wherein the trained neural network classifies the compartment into the first category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint does not compare to previously classified single images of known prohibited objects known by the trained neural network.

5. The method of claim 3 , wherein the trained neural network classifies the compartment into the second category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint compares to previously classified single images of known prohibited objects known by the trained neural network.

6. The method of claim 3 , wherein the calculating the maximum intensity projection across each dimension of at least two orthogonal projections of the three-dimensional representation includes:

first calculating, by the processor, a maximum value of a projection in a Z dimension along an X, Y plane of the at least two orthogonal projections, for each pixel;

second calculating, by the processor, a maximum value of a projection in a Y dimension along a X, Z plane of the at least two orthogonal projections, for each pixel; and

third calculating, by the processor, a maximum value of a projection in an X dimension along a Y, Z plane of the at least two orthogonal projections, for each pixel.

7. The method of claim 1 , wherein the trained neural network is a two-dimensional convolutional neural network (CNN) that classifies the compartment into the first category or the second category.

8. The method of claim 1 , wherein the first category represents compartments that do not include a prohibited object based on an output of the trained neural network, and the second category represents compartments that do include a prohibited object based on the output of the trained neural network.

9. A computing system, comprising:

a processor;

a memory device coupled to the processor; and

a computer readable storage device coupled to the processor, wherein the storage device contains program code executable by the processor via the memory device to implement a method for using augmented reality for automatically classifying compartments at a security checkpoint comprising:

receiving, by a processor of a computing system, a three-dimensional representation of a compartment from an imaging device connected to the computing system; and

classifying, by the processor, the compartment into a first category or a second category using a trained neural network that analyzes the three-dimensional representation, wherein compartments classified into the first category are automatically cleared at the security checkpoint and compartments classified into the second category undergo further security checks.

10. The computing system of claim 9 , wherein the three-dimensional representation is a set of images in sequence of the interior region of the compartment, output by a computed tomography (CT) scanner, further wherein each image in the set of images has a width and a height, and when stacked together, form a depth.

11. The computing system of claim 9 , wherein the trained neural network is trained by:

calculating, by the processor, a maximum intensity projection across each dimension of at least two orthogonal projections of a known three-dimensional representation of a prohibited object;

stacking, by the processor, each of the maximum intensity projections into a single image; and

feeding, by the processor, the single image into the trained neural network so that the trained neural network learns that the single image having the same maximum intensity projections corresponds to the prohibited object.

12. The computing system of claim 11 , wherein the trained neural network classifies the compartment into the first category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint does not compare to previously classified single images of known prohibited objects known by the trained neural network.

13. The computing system of claim 11 , wherein the trained neural network classifies the compartment into the second category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint compares to previously classified single images of known prohibited objects known by the trained neural network.

14. The computing system of claim 11 , wherein the calculating the maximum intensity projection across each dimension of at least two orthogonal projections of the three-dimensional representation includes:

first calculating, by the processor, a maximum value of a projection in a Z dimension along an X, Y plane of the at least two orthogonal projections, for each pixel;

second calculating, by the processor, a maximum value of a projection in a Y dimension along a X, Z plane of the at least two orthogonal projections, for each pixel; and

third calculating, by the processor, a maximum value of a projection in an X dimension along a Y, Z plane of the at least two orthogonal projections, for each pixel.

15. The computing system of claim 9 , wherein the trained neural network is a two-dimensional convolutional neural network (CNN) that classifies the compartment into the first category or the second category regardless of a shape or a material of an object contained within the compartment, without requiring human inspection of an output of the imaging device.

16. The computing system of claim 9 , wherein the first category represents compartments that do not include a prohibited object based on an output of the trained neural network, and the second category represents compartments that do include a prohibited object based on the output of the trained neural network.

17. A computer program product, comprising a computer readable hardware storage device storing a computer readable program code, the computer readable program code comprising an algorithm that when executed by a computer processor of a computing system implements a method for automatically classifying compartments at a security checkpoint comprising:

receiving, by the processor, a three-dimensional representation of a compartment from an imaging device connected to the computing system; and

classifying, by the processor, the compartment into a first category or a second category using a trained neural network that analyzes the three-dimensional representation, regardless of a shape or a material of an object contained within the compartment, without requiring human inspection of an output of the imaging device, wherein compartments classified into the first category are automatically cleared at the security checkpoint and compartments classified into the second category undergo further security checks.

18. The computer program product of claim 17 , wherein the trained neural network is trained by:

calculating, by the processor, a maximum intensity projection across each dimension of at least two orthogonal projections of a known three-dimensional representation of a prohibited object;

stacking, by the processor, each of the maximum intensity projections into a single image; and

feeding, by the processor, the single image into the trained neural network so that the trained neural network learns that the single image having the same maximum intensity projections corresponds to the prohibited object.

19. The computer program product of claim 18 , wherein:

the trained neural network classifies the compartment into the first category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint does not compare to previously classified single images of known prohibited objects known by the trained neural network; and

the trained neural network classifies the compartment into the second category in response to determining that a single image fed to the trained neural network calculated by the computing system from the three-dimensional representation of the compartment being scanned at the security checkpoint compares to previously classified single images of known prohibited objects known by the trained neural network.

20. The computer program product of claim 18 , wherein the calculating the maximum intensity projection across each dimension of at least two orthogonal projections of the three-dimensional representation includes:

first calculating, by the processor, a maximum value of a projection in a Z dimension along an X, Y plane of the at least two orthogonal projections, for each pixel;

second calculating, by the processor, a maximum value of a projection in a Y dimension along a X, Z plane of the at least two orthogonal projections, for each pixel; and

third calculating, by the processor, a maximum value of a projection in an X dimension along a Y, Z plane of the at least two orthogonal projections, for each pixel.