# US 2022/0044387 A1 · Systems and methods for color agnostic material face detection

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

- Number: US 2022/0044387 A1
- Document ID: US20220044387A1
- Status: Published application
- Kind: A1
- Application number: 16/987,638
- Filed: 2020-08-07
- Published: 2022-02-10
- Earliest claimed: 2020-08-07
- Assignees: SoftWear Automation, Inc.
- Applicants: SoftWear Automation Inc.
- Inventors: Michael J. Baker; Alexander Ren Gurney; Senthil Ramamurthy
- Web page: https://patentgrep.com/patent/US20220044387A1

## CPC

- G06T 7/0004: {Industrial image inspection}
- G06V 10/40: Extraction of image or video features
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06T 2207/10024: Color image
- G06T 2207/20081: Training; Learning
- G06T 2207/30124: Fabrics; Textile; Paper
- G06V 2201/06: Recognition of objects for industrial automation
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

Various examples are provided related to face identification of material. An image can be captured by a vision system and feature parameters determined and compared to a material feature database to determine which face of the material is being presented. The vision system can employ a set of parameters for configuration to acquire the image. The system can communicate the identification to downstream processes in real time. A near-universal, color agnostic, angular orientation independent identification of material faces can be determined without the need for physical manipulation of the material.

## Claims

### Claim 1

A material face identification system, comprising:
a vision system comprising at least one optical sensor configured to obtain images of a work area;
a face identification module; and
processing circuitry comprising a processor, wherein execution of the face identification module by the processor causes the processing circuitry to:
obtain imaging parameters associated with a piece of material provided to the work area, the piece of material having a first side exposed to the at least one optical sensor and a second side opposite the first side;
initiate capture of an image of at least a portion of the exposed first side of the piece of material by the vision system, where the vision system is configured to capture the image with the at least one optical sensor based upon the imaging parameters; and
determine whether the exposed first side is a correct face of the piece of material that is facing upward in the work area based upon a confidence score generated using feature data determined from the image of the exposed first side of the piece of material.

### Claim 2 (depends on 1)

The material face identification system of claim 1, wherein a machine learning model associated with the piece of material generates the confidence score based upon the feature data extracted from the image of the exposed first side of the piece of material.

### Claim 3 (depends on 2)

The material face identification system of claim 2, wherein the feature data comprises a compact feature set generated from material features of the piece of material.

### Claim 4 (depends on 3)

The material face identification system of claim 3, wherein the material features are extracted from the image of the exposed first side of the piece of material.

### Claim 5 (depends on 4)

The material face identification system of claim 4, wherein the material features comprise filtered channel response that are devoid of color information.

### Claim 6 (depends on 4)

The material face identification system of claim 4, wherein the material features are based at least in part upon material texture.

### Claim 7 (depends on 2)

The material face identification system of claim 2, wherein a second confidence score is generated for a second side of the piece of material, where the correct face is determined at least in part upon the confidence scores.

### Claim 8 (depends on 1)

The material face identification system of claim 1, wherein execution of the face identification module determines if the piece of material is present in the work area.

### Claim 9 (depends on 1)

The material face identification system of claim 1, wherein the imaging parameters associated with the piece of material are retrieved from an imaging parameter database based upon a material ID associated with the piece of material.

### Claim 10 (depends on 9)

The material face identification system of claim 9, wherein a machine learning model is selected based upon the material ID, the machine learning model trained to generate the confidence score based upon the feature data determined from the image of the piece of material.

### Claim 11 (depends on 1)

The material face identification system of claim 1, wherein the piece of material is flipped to expose a second face of the piece of material in response to the correct face not facing upward.

### Claim 12

(canceled)

### Claim 13 (depends on 1)

The material face identification system of claim 1, wherein the imaging parameters are determined based upon quality scores generated for a plurality of test images of a face of at least one sample of the material, the plurality of test images captured with different combinations of imaging parameters.

### Claim 14 (depends on 13)

The material face identification system of claim 13, wherein the imaging parameters corresponding to the quality score that is highest are used to capture the image of the piece of material.

### Claim 15 (depends on 2)

The material face identification system of claim 2, wherein the machine learning model is trained based upon material features determined from at least one sample image of a face of at least one sample of the material.

### Claim 16 (depends on 15)

The material face identification system of claim 15, wherein a compact feature set generated from the material features is used to train the machine learning model corresponding to the at least one sample of the material.

### Claim 17 (depends on 15)

The material face identification system of claim 15, wherein the at least one sample image comprises images obtained at different locations on the at least one sample of the material.

### Claim 18 (depends on 17)

The material face identification system of claim 17, wherein the machine learning model is trained for both faces of the at least one sample of the material.

### Claim 19 (depends on 15)

The material face identification system of claim 15, wherein the material features comprise texture features from at least one image of the at least one sample of the material.

### Claim 20 (depends on 15)

The material face identification system of claim 15, wherein the material features comprise filtered channel response that are devoid of color information.
