Cognitive screening for prohibited items across multiple checkpoints by using context aware spatio-temporal analysis

1. A computer-implemented screening method, the method comprising:

detecting a disassembled component of an object in a first sensor scan image of a first container belonging to a first owner and another disassembled component of the object in a second sensor scan image of a second container belonging to a second owner, the first sensor scan image being linked to the first owner and the second sensor scan image being linked to the second owner, the detecting being based on a trained model that is run both offline and online;

storing a record of the disassembled component correlated with contextual information of the first owner, and storing a record of the another disassembled component correlated with contextual information of the second owner, the contextual information comprising a departure location and a destination location of the respective owners;

determining that the disassembled component can be combined with the another disassembled component to assemble the object;

determining a degree of danger posed by assembly of the object;

determining a correlation between the first owner and the second owner based on the contextual information of the respective owners and a past relationship between the first owner and the second owner, the correlation indicating a likelihood of assembly of the object; and

determining whether to flag the first owner and the second owner based on an assessment of the degree of danger and the correlation, wherein the determining whether to flag comprises:

when the correlation is greater than a linking threshold, flagging the first owner and the second owner;

when the correlation is less than the linking threshold and the degree of danger is less than a threat threshold, no flagging is performed; and

when the correlation is less than the linking threshold and the degree of danger is greater than the threat threshold, flagging the first owner and the second owner.

2. The computer-implemented method of claim 1 , wherein the sensor scan images comprise sensor scan images that are at least one of spatially and temporally disparate, and

wherein the detecting uses a Deformable Parts Model technique for a detection technique.

3. The computer-implement method of claim 1 , wherein the storing stores the record of each disassembled component in a configurable database that is configured based on a time period of relevant sensor scan images.

4. The computer-implement method of claim 1 , wherein the flagging flags the disassembled component irrespective of a prior sensor scan image including the another disassembled component if a threat of the object is greater than a predetermined threshold.

5. The method of claim 1 , embodied in a cloud-computing environment.

6. A computer program product, the computer program product comprising a non-transitory computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform:

detecting a disassembled component of an object in a first sensor scan image of a first container belonging to a first owner and another disassembled component of the object in a second sensor scan image of a second container belonging to a second owner, the first sensor scan image being linked to the first owner and the second sensor scan image being linked to the second owner, the detecting being based on a trained model that is run both offline and online;

storing a record of the disassembled component correlated with contextual information of the first owner, and storing a record of the another disassembled component correlated with contextual information of the second owner, the contextual information comprising a departure location and a destination location of the respective owners;

determining that the disassembled component can be combined with the another disassembled component to assemble the object;

determining a degree of danger posed by assembly of the object;

determining a correlation between the first owner and the second owner based on the contextual information of the respective owners and a past relationship between the first owner and the second owner, the correlation indicating a likelihood of assembly of the object; and

determining whether to flag the first owner and the second owner based on an assessment of the degree of danger and the correlation, wherein the determining whether to flag comprises:

when the correlation is greater than a linking threshold, flagging the first owner and the second owner;

when the correlation is less than the linking threshold and the degree of danger is less than a threat threshold, no flagging is performed; and

when the correlation is less than the linking threshold and the degree of danger is greater than the threat threshold, flagging the first owner and the second owner.

7. The computer program product of claim 6 , wherein the storing stores the record of each disassembled component in a configurable database that is configured based on a time period of relevant sensor scan images.

8. The computer program product of claim 6 , wherein the flagging flags the disassembled component irrespective of a prior sensor scan image including the another disassembled component if a threat of the object is greater than a predetermined threshold.

9. A prediction system, the system comprising:

a processor; and

a memory operably coupled to the processor, the memory storing instructions to cause the processor to perform:

detecting a disassembled component of an object in a first sensor scan image of a first container belonging to a first owner and another disassembled component of the object in a second sensor scan image of a second container belonging to a second owner, the first sensor scan image being linked to the first owner and the second sensor scan image being linked to the second owner, the detecting being based on a trained model that is run both offline and online;

storing a record of the disassembled component correlated with contextual information of the first owner, and storing a record of the another disassembled component correlated with contextual information of the second owner, the contextual information comprising a departure location and a destination location of the respective owners;

determining that the disassembled component can be combined with the another disassembled component to assemble the object;

determining a degree of danger posed by assembly of the object;

determining a correlation between the first owner and the second owner based on the contextual information of the respective owners and a past relationship between the first owner and the second owner, the correlation indicating a likelihood of assembly of the object; and

determining whether to flag the first owner and the second owner based on an assessment of the degree of danger and the correlation, wherein the determining whether to flag comprises:

when the correlation is greater than a linking threshold, flagging the first owner and the second owner;

when the correlation is less than the linking threshold and the degree of danger is less than a threat threshold, no flagging is performed; and

when the correlation is less than the linking threshold and the degree of danger is greater than the threat threshold, flagging the first owner and the second owner.

10. The system of claim 9 , embodied in a cloud-computing environment.