Systems and methods for image processing
AssigneeL-3 Security & Detection Systems, Inc.
InventorsDavid Perticone, Andrew D. Foland
Life of this patent
Estimating the term…
Abstract
A computing-device implemented system and method for identifying an item in an x-ray image is described. The method includes training a machine learning algorithm with at least one training data set of x-ray images to generate at least one machine-learned model. The method further includes receiving at least one rendered x-ray image that includes an item, identifying the item using the at least one model, and generating an automated detection indication associated with the item.
Claims51
A computing-device implemented method for identifying an item in an x-ray image using at least one computing device equipped with a processor, the method comprising:
- executing, using the at least one computing device, at least one machine-learned model, the at least one machine-learned model trained with at least one training data set of x-ray images, the at least one training data set including a first set of x-ray images of items containing threats and a second set of x-ray images of items not containing threats, the at least one machine-learned model is a convolutional neural network (CNN) and the first set of x-ray images or the second set of x-ray images include one or more volumetric x-ray images or one or more projection x-ray images;
- receiving, with the at least one computing device, at least one rendered x-ray image that includes an item, the x-ray image having a 2-D TRI-MAT format that identifies items in the image as organic, inorganic or metallic;
- identifying, with the at least one computing device, the item using the at least one model;
- and
- generating, with the at least one computing device, an automated detection indication associated with the item.
A system for identifying an item in an x-ray image, the system comprising:
- a scanner configured to render at least one x-ray image;
- a computing device equipped with a processor in communication with the scanner, the computing device configured to:
- execute at least one machine-learned model, the at least one machine-learned model trained with at least one training data set of x-ray images, the at least one training data set including a first set of x-ray images of items containing threats and a second set of x-ray images of items not containing threats, the at least one machine-learned model being is a convolutional neural network (CNN), and the first set of x-ray images or the second set of x-ray images include one or more volumetric x-ray images or one or more projection x-ray images;
- receive the at least one rendered x-ray image from the scanner, the at least one rendered x-ray image includes an item, the x-ray image having a 2-D TRI-MAT format that identifies items in the image as organic, inorganic or metallic;
- identify the item using the at least one model;
- and
- generate an automated detection indication associated with the item.
The system of 19, wherein the at least one rendered x-ray image is based on scanning at least one object at a security checkpoint in a facility.
A non-transitory computer readable medium storing instructions executable by a processor, wherein execution of the instructions causes the processor to implement a method for identifying an item in an x-ray image, the method comprising:
- executing at least one machine-learned model, the at least one machine-learned model trained with at least one training data set of x-ray images, the at least one training data set including a first set of x-ray images of items containing threats and a second set of x-ray images of items not containing threats, the at least one machine-learned model is a convolutional neural network (CNN) and the first set of x-ray images or the second set of x-ray images include one or more volumetric x-ray images or one or more projection x-ray images;
- receiving at least one rendered x-ray image that includes an item, the x-ray image having a 2-D TRI-MAT format that identifies items in the image as organic, inorganic or metallic;
- identifying the item using the at least one model;
- and
- generating an automated detection indication associated with the item.
The non-transitory computer readable medium of 35, wherein the at least one training data set includes a first set of x-ray images containing a specified item and second set of x-ray images not containing the specified item.
The non-transitory computer readable medium of 35, wherein the automated detection indication is based at least in part on machine-learned features of the item that are based on the at least one training data set.
The non-transitory computer readable medium of 35, wherein the machine-learned model includes weights, coefficients, and forms learned by analysis of the at least one training data set.
Dependencies are inferred from the claim text.