# US 10,872,409 B2 · Visual augmentation of regions within images

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,872,409 B2
- Document ID: US10872409B2
- Status: Granted
- Kind: B2
- Application number: 15/890,435
- Filed: 2018-02-07
- Published: 2019-08-08
- Granted: 2020-12-22
- Earliest claimed: 2018-02-07
- Assignees: Analogic Corporation
- Applicants: Analogic Corporation
- Inventors: David Schafer
- Web page: https://patentgrep.com/patent/US10872409B2

## CPC

- G06T 7/0012: {Biomedical image inspection}
- G06T 7/11: Region-based segmentation
- G06T 7/155: involving morphological operators
- G06T 7/187: involving region growing; involving region merging; involving connected component labelling
- G06T 2200/04: involving 3D image data
- G06T 2207/10076: 4D tomography; Time-sequential 3D tomography
- G06T 2207/10081: Computed x-ray tomography [CT]
- G06T 2207/20076: Probabilistic image processing
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30112: Baggage; Luggage; Suitcase
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Among other things, one or more systems and/or techniques for visually augmenting regions within images are provided herein. An image of an object, such as a bag, is segmented to identify an item (e.g., a metal gun barrel). Features of the item are extracted from voxels representing the item within the image (e.g., voxels within a first region), such as a size, shape, density, and orientation of the item. Response to the features of the item matching predefined features of a target item to detect, one or more additional regions are identified, such as a second region proximate to the first region based upon a location of the second region corresponding to where a connected part of the item (e.g., a plastic handle of the gun) is predicted to be located. The one or more regions are visually distinguished within the image from other regions (e.g., colored, highlighted, etc.).

## Claims

### Claim 1

A method for visually augmenting regions within an image of an object, comprising:
segmenting the image of the object to identify an item depicted within a first region of the image;
extracting features of the item from the image;
comparing the features of the item to predefined features of a target item to be detected;
responsive to determining that the features of the item match the predefined features within a threshold amount, identifying a second region proximate to the first region comprising the item based upon a location of the second region corresponding to where a connected part of the item is predicted to be located;
applying a visual augmentation to the second region to visually distinguish the second region from other regions depicted within the image to create an augmented image; and
rendering the augmented image through a display.

### Claim 2 (depends on 1)

The method of claim 1, comprising:
performing segmentation and feature extraction upon the second region to identify the connected part; and
applying the visual augmentation to the connected part depicted within the second region.

### Claim 3 (depends on 1)

The method of claim 1, wherein the identifying a second region comprises:
identifying a size, a shape, and an orientation of the item; and
predicting where the connected part would be located with respect to the item based upon the size, the shape, and the orientation of the item.

### Claim 4 (depends on 1)

The method of claim 1, wherein the identifying a second region comprises:
identifying an axis of the item; and
defining the second region as being located at an end of the axis orthogonal to the item.

### Claim 5 (depends on 1)

The method of claim 1, wherein the identifying a second region comprises:
evaluating density of voxels within the second region to identify a set of voxels with densities within a threshold similarity; and
identifying the connected part as being represented by the set of voxels.

### Claim 6 (depends on 1)

The method of claim 1, wherein the item comprises a material with a larger density than the connected part.

### Claim 7 (depends on 1)

The method of claim 1, comprising:
performing density thresholding upon the image to zero out voxels having densities below a threshold.

### Claim 8

A computing device tomography system, comprising:
a radiation source;
a detector array;
an image reconstructor configured to generate a three-dimensional image of an object; and
an item detection component configured to:
identify an item within a first region of the three-dimensional image as a potential item of interest by comparing features of items depicted within the three-dimensional image to predefined features of an item of interest to be detected;
responsive to determining that features of the item match the predefined features within a threshold amount, identify a second region proximate to the first region comprising the item based upon a location of the second region corresponding to where a connected part of the item is predicted to be located;
apply a visual augmentation to the second region to visually distinguish the second region from other regions depicted within the image to create an augmented image; and
render the augmented image through a display.

### Claim 9 (depends on 8)

The computing device of claim 8, wherein the item detection component is configured to:
apply a color augmentation to the second region.

### Claim 10 (depends on 8)

The computing device of claim 8, wherein the item detection component is configured to:
apply a contrast augmentation to the second region.

### Claim 11 (depends on 8)

The computing device of claim 8, wherein the item detection component is configured to:
apply an edge enhancement augmentation to the second region.

### Claim 12 (depends on 8)

The computing device of claim 8, wherein the item detection component is configured to:
determine a predicted location of the connected part within the second region; and
apply the visual augmentation to the connected part at the predicted location.

### Claim 13 (depends on 8)

The computing device of claim 8, wherein the item detection component is configured to:
determine a context associated with the object; and
adjust a value of the threshold amount based upon the context.

### Claim 14 (depends on 13)

The computing device of claim 13, wherein the context corresponds to a current threat level.

### Claim 15 (depends on 13)

The computing device of claim 13, wherein the context corresponds to information regarding an owner of the object.

### Claim 16

A non-transitory computer-readable medium comprising computer-executable instructions that when executed via a processing unit cause performance of operations, the operations comprising:
identifying a first item depicted within a first region of an image of an object based upon the first item having a density above a threshold value;
determining that the first item has a probability equal to at least a threshold amount of being part of a target item to be detected;
selecting a second region proximate to the first region, the second region selected as a function of an orientation of the first item;
evaluating the second region to determine whether a second item is present within the second region; and
upon determining that the second item is present within the second region, grouping voxels of the first item with voxels of the second item such that the first item and the second item are treated as a single item.

### Claim 17 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the determining that the first item has a probability equal to at least a threshold amount of being part of a target item comprises:
comparing morphological characteristics of the first item to known characteristics of the target item.

### Claim 18 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the operations comprise:
performing object detection on the grouped voxels of the first item and the second item to determine whether the single item is a potential threat item.

### Claim 19 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the operations comprise:
segmenting the image using a region growing technique by clustering voxels based upon density similarity.

### Claim 20 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the operations comprise:
segmenting the image using a connected components technique by scanning across the image to connect voxels with similar densities.