Systems and methods for image processing
AssigneeL3 Security & Detection Systems, Inc.
InventorDenis Dujmic
Life of this patent
Estimating the term…
Abstract
Systems and methods for identifying objects in shipping containers using radiographic images are described. More particularly, techniques for using artificial neural networks to analyze radiographic images of shipping containers and associated cargo manifests are discussed.
Claims50
A computing-device implemented method for validating a cargo manifest associated with a shipping container, the method comprising:
- receiving, at a computing device, a rendered radiographic image of a scanned shipping container from a scanner configured to scan the shipping container and render the radiographic image;
- extracting, with an autoencoder neural network, a feature vector from the radiographic image;
- assigning a statistical probability that an object in the scanned shipping container matches a listed object on an associated cargo manifest for the scanned shipping container through comparison of the extracted feature vector against an historic distribution of prior feature vectors of radiographic images of prior shipping containers associated with cargo manifests containing a like object;
- and
- generating an automated detection indication related to the scanned shipping container based on the assigned statistical probability.
-22. (canceled)
A system for validating a cargo manifest associated with a shipping container, the system comprising:
- a scanner configured to scan a shipping container and render at least one radiographic image;
- a computing device equipped with one or more processors and communicatively coupled to the scanner, the computing device configured to:
- receive a rendered radiographic image of a scanned shipping container from a scanner configured to scan the shipping container and render the radiographic image,
- extract, with an autoencoder neural network, a feature vector from the radiographic image,
- assign a statistical probability that an object in the scanned shipping container matches a listed object on an associated cargo manifest for the scanned shipping container through comparison of the extracted feature vector against an historic distribution of prior feature vectors of radiographic images of prior shipping containers associated with cargo manifests containing a like object, and
- generate an automated detection indication related to the scanned shipping container based on the assigned statistical probability;
- and
- an output mechanism configured to display graphics or generate audio based on the automated detection indication.
-60. (canceled)
A non-transitory computer readable medium storing instructions executable by a processor for validating a cargo manifest associated with a shipping container, wherein execution of the instructions causes a computing device to:
- receive a rendered radiographic image of a scanned shipping container from a scanner configured to scan the shipping container and render the radiographic image;
- extract, with an autoencoder neural network, a feature vector from the radiographic image;
- assign a statistical probability that an object in the scanned shipping container matches a listed object on an associated cargo manifest for the scanned shipping container through comparison of the extracted feature vector against a historic distribution of prior feature vectors of radiographic images of prior shipping containers associated with cargo manifests containing a like object;
- and
- generate an automated detection indication related to the scanned shipping container based on the assigned statistical probability.
-83. (canceled)
Dependencies are inferred from the claim text.