Back to results
US 11,010,605 B2Granted
Multi-model detection of objects
AssigneeRapiscan Laboratories, Inc.
InventorsClaire Nord, Simanta Gautam, Bruno Brasil Ferrari Faviero, Steven Posun Yang, Jay Harshadbhai Patel, Ian Cinnamon
Life of this patent
Estimating the term…
Abstract
Disclosed is an object-detection system configured to utilize multiple object-detection models that generate respective sets of object-detection conclusions to detect objects of interest within images of scenes. The object-detection system is configured to implement a series of functions to reconcile any discrepancies that exist in its multiple sets of object-detection conclusions in order to generate one set of conclusions for each perceived object of interest within a given image.
Claims20
A computing system comprising:
- at least one processor;
- a non-transitory computer-readable medium;
- and
- program instructions stored on the non-transitory computer-readable medium that are executable by the at least one processor to cause the computing system to:
- receive an image of a scene that was scanned by a detection device of a screening system, wherein the image comprises an object of interest;
- identify at least a first set of object-detection conclusions for the image generated by a first object-detection model and a second set of object-detection conclusions for the image generated by a second object-detection model, wherein at least one conclusion discrepancy exists between the first and second set of object-detection conclusions;
- resolve the at least one conclusion discrepancy by:
- determining a measure of similarity between a first localization conclusion from the first set of object-detection conclusions and a second localization conclusion from the second set of object-detection conclusions, wherein the first and second localization conclusions were both determined to refer to the object of interest;
- defining a plurality of candidate partitions, wherein at least one candidate partition of the plurality of candidate partitions is defined based on the measure of similarity between the first and second localization conclusions;
- and
- based on a weight assigned to the at least one candidate partition, deriving a reconciled set of object-detection conclusions for the object of interest from the first and second set of object-detection conclusions, thereby resolving the at least one conclusion discrepancy;
- and
- based on the reconciled set of object-detection conclusions, determine that the image of the scene comprises the object of interest.
10 more elements
A tangible, non-transitory computer-readable medium comprising program instructions stored thereon that are executable by at least one processor to cause a computing system to:
- receive an image of a scene that was scanned by a detection device of a screening system, wherein the image comprises an object of interest;
- identify at least a first set of object-detection conclusions for the image generated by a first object-detection model and a second set of object-detection conclusions for the image generated by a second object-detection model, wherein at least one conclusion discrepancy exists between the first and second set of object-detection conclusions;
- resolve the at least one conclusion discrepancy by:
- determining a measure of similarity between a first localization conclusion from the first set of object-detection conclusions and a second localization conclusion from the second set of object-detection conclusions, wherein the first and second localization conclusions were both determined to refer to the object of interest;
- defining a plurality of candidate partitions, wherein at least one candidate partition of the plurality of candidate partitions is defined based on the measure of similarity between the first and second localization conclusions;
- and
- based on a weight assigned to the at least one candidate partition, deriving a reconciled set of object-detection conclusions for the object of interest from the first and second set of object-detection conclusions, thereby resolving the at least one conclusion discrepancy;
- and
- based on the reconciled set of object-detection conclusions, determine that the image of the scene comprises the object of interest.
6 more elements
A computer-implemented method comprising:
- receiving an image of a scene that was scanned by a detection device of a screening system, wherein the image comprises an object of interest;
- identifying at least a first set of object-detection conclusions for the image generated by a first object-detection model and a second set of object-detection conclusions for the image generated by a second object-detection model, wherein at least one conclusion discrepancy exists between the first and second set of object-detection conclusions;
- resolving the at least one conclusion discrepancy by:
- determining a measure of similarity between a first localization conclusion from the first set of object-detection conclusions and a second localization conclusion from the second set of object-detection conclusions, wherein the first and second localization conclusions were both determined to refer to the object of interest;
- defining a plurality of candidate partitions, wherein at least one candidate partition of the plurality of candidate partitions is defined based on the measure of similarity between the first and second localization conclusions;
- and
- based on a weight assigned to the at least one candidate partition, deriving a reconciled set of object-detection conclusions for the object of interest from the first and second set of object-detection conclusions, thereby resolving the at least one conclusion discrepancy;
- and
- based on the reconciled set of object-detection conclusions, determining that the image of the scene comprises the object of interest.
6 more elements
Dependencies are inferred from the claim text.