# US 2026/0065648 A1 · Imaging System and Method for Deploying Greedy Optimization for Training Machine Vision

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

- Number: US 2026/0065648 A1
- Document ID: US20260065648A1
- Status: Published application
- Kind: A1
- Application number: 18/821,438
- Filed: 2024-08-30
- Published: 2026-03-05
- Earliest claimed: 2024-08-30
- Assignees: Zebra Technologies Corporation
- Applicants: Zebra Technologies Corporation
- Inventors: Caelan Marks; Dominique Rivard; Andre Al-Khoury; Pankaj R. Roy
- Web page: https://patentgrep.com/patent/US20260065648A1

## CPC

- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06T 7/0004: {Industrial image inspection}
- G06V 10/771: Feature selection, e.g. selecting representative features from a multi-dimensional feature space
- G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods}
- G06V 10/82: using neural networks
- G06T 2207/20016: Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

Systems and methods are provided for training an imaging system for anomaly detection through a machine learning architecture combined with a local memory optimizing, iterative sub-sampling process. Training includes separating training images into subsets and iteratively feeding each subset to the machine learning architecture which extracts patch-level features in a feature space. A sub-sampling process generates a coreset from these extracted features, each iteration. Each iteration new extracted features and the existing coreset and are fed to the sub-sampling process which updates the coreset, iteratively until all training images are consumed. To aid optimization, each iteration a convergence value indicating coreset generation progress is determined for displaying status of the anomaly detection training.

## Claims

### Claim 1

A method of training an imaging device for anomaly detection, the method comprising:
in a training mode, receiving, at one or more processors, a set of training images of an object;
separating the set of training images into n subsets of the training images, where n is an integer greater than 1;
iteratively, for each subset of training images,
feeding the subset of training images to a machine learning framework trained to extract patch-level features in a feature space, each patch-level feature corresponding to a different location in the subset of training images,
extracting, using the machine learning framework, patch-level features for the subset of training images, and feeding the extracted patch-level features and, in response to a coreset of features being stored in a memory, feeding the coreset of features, to a sub-sampling algorithm,
generating, in the sub-sampling algorithm, an updated coreset of features, and
generating, at the sub-sampling algorithm, convergence values corresponding to a performance metric of the sub-sampling algorithm; and
storing the updated coreset of features for use in anomaly detection on subsequently captured images of the object during an inference mode.

### Claim 2 (depends on 1)

The method of claim 1, wherein the sub-sampling algorithm is a patch-level feature selection algorithm and the convergence values corresponds to a distance metric determined by the sub-sampling algorithm.

### Claim 3 (depends on 2)

The method of claim 2, the method further comprising generating a graphical display of the convergence values for each iteration for display to a user.

### Claim 4 (depends on 1)

The method of claim 1, wherein the sub-sampling algorithm comprises k-center clustering, grid sampling, furthest point sampling, statistical sampling, or random sampling.

### Claim 5 (depends on 1)

The method of claim 1, wherein the machine learning framework trained to extract the patch-level features is a convolutional neural network.

### Claim 6 (depends on 1)

The method of claim 1, wherein the set of training images of the object comprise whole images of the object.

### Claim 7 (depends on 1)

The method of claim 1, wherein the set of training images of the object comprise tile images of the object.

### Claim 8 (depends on 1)

The method of claim 1, wherein the set of training images of the object comprise images of the object at different scales.

### Claim 9

An imaging system comprising:
an imaging device configured to capture images of an object;
a processor, a memory, and a computer-readable media storage having machine readable instructions stored thereon that, when the machine readable instructions are executed, cause the imaging system to:
receive, at the processor, a set of training images of an object;
separate the set of training images into n subsets of the training images, where n is an integer greater than 1;
iteratively, for each subset of training images,
feed the subset of training images to a machine learning framework trained to extract patch-level features in a feature space, each patch-level feature corresponding to a different location in the subset of training images,
extract, using the machine learning framework, patch-level features for the subset of training images, and feed the extracted patch-level features and, in response to a coreset of features being stored in the memory, feed the coreset of features, to a sub-sampling algorithm,
generate, in the sub-sampling algorithm, an updated coreset of features, and
generate, at the sub-sampling algorithm, convergence values corresponding to a performance metric of the sub-sampling algorithm; and
store the updated coreset of features for use in anomaly detection on subsequently captured images of the object during an inference mode.

### Claim 10 (depends on 9)

The imaging system of claim 9, wherein the sub-sampling algorithm is a patch-level feature selection algorithm and the convergence values corresponds to a distance metric determined by the sub-sampling algorithm.

### Claim 11 (depends on 9)

The imaging system of claim 9, wherein the machine readable instructions include further instructions that, when executed, cause the imaging system to generate a graphical display of the convergence values for each iteration for display to a user.

### Claim 12 (depends on 9)

The imaging system of claim 9, wherein the sub-sampling algorithm comprises k-center clustering, grid sampling, furthest point sampling, statistical sampling, or random sampling.

### Claim 13 (depends on 9)

The imaging system of claim 9, wherein the machine learning framework trained to extract the patch-level features is a convolutional neural network.

### Claim 14 (depends on 9)

The imaging system of claim 9, wherein the set of training images of the object comprise whole images of the object.

### Claim 15 (depends on 9)

The imaging system of claim 9, wherein the set of training images of the object comprise tile images of the object.

### Claim 16 (depends on 9)

The imaging system of claim 9, wherein the set of training images of the object comprise images of the object at different scales.
