# US 2019/0056781 A1 · Eye tracking apparatus and method thereof

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

- Number: US 2019/0056781 A1
- Document ID: US20190056781A1
- Status: Published application
- Kind: A1
- Application number: 15/846,209
- Filed: 2017-12-19
- Published: 2019-02-21
- Earliest claimed: 2017-08-16
- Assignees: Industrial Technology Research Institute
- Applicants: Industrial Technology Research Institute
- Inventors: Hsiao-Wei Liu; Yu-Ying Lan; Hsin-Cheng Lin; Chung-Lun Kuo; Chia-Liang Yeh
- Web page: https://patentgrep.com/patent/US20190056781A1

## CPC

- G06F 3/013: {Eye tracking input arrangements (G06F3/015 takes precedence)}
- G06T 7/75: {involving models}
- G06V 40/193: {Preprocessing; Feature extraction}
- G06V 40/197: {Matching; Classification}
- G02B 27/0093: {with means for monitoring data relating to the user, e.g. head-tracking, eye-tracking}
- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06T 7/73: using feature-based methods
- G06T 7/74: {involving reference images or patches}
- G06T 2207/30041: Eye; Retina; Ophthalmic
- G06T 2207/30201: Face
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G02B: OPTICAL ELEMENTS, SYSTEMS OR APPARATUS

## Abstract

An eye tracking apparatus and an eye tracking method for the eye tracking apparatus are provided. The eye tracking apparatus includes an image capture device and a computing device. The image capture device is adapted to capture at least one face image. The computing device is coupled to the image capture device. The computing device receives the face image to obtain an eye image and identifies at least one iris region of the eye image. The computing device chooses the largest iris region from the iris regions to obtain a fitting pattern of the largest iris region and obtains a gaze point of the eye image based on the fitting pattern.

## Claims

### Claim 1

An eye tracking apparatus, comprising:
an image capture device, configured to capture at least one face image; and
a computing device, coupled to the image capture device, receiving the face image to obtain an eye image, identifying at least one iris region in the eye image, and choosing a largest iris region from the at least one iris region and obtaining a fitting pattern of the largest iris region, and obtaining a gaze point of the eye image based on the fitting pattern.

### Claim 2 (depends on 1)

The eye tracking apparatus as claimed in claim 1, wherein the fitting pattern is a minimum covering circle or a minimum covering ellipse.

### Claim 3 (depends on 1)

The eye tracking apparatus as claimed in claim 1, wherein the computing device performs an image enhancing operation to enhance the eye image and thereby generate an enhanced eye image, and defines the at least one iris region in the enhanced eye image.

### Claim 4 (depends on 3)

The eye tracking apparatus as claimed in claim 3, wherein the image enhancing operation comprises an image brightness adjustment operation, an image contrast increment operation and/or a Weickert image enhancing operation.

### Claim 5 (depends on 4)

The eye tracking apparatus as claimed in claim 4, wherein the image brightness adjustment comprises increasing a brightness at a position with an excessively low grayscale value in the eye image.

### Claim 6 (depends on 4)

The eye tracking apparatus as claimed in claim 4, wherein the image brightness adjustment comprises lowering a brightness at a position with an excessively high grayscale value in the eye image.

### Claim 7 (depends on 3)

The eye tracking apparatus as claimed in claim 3, further comprising:
a training device, coupled to the computing device and configured to perform a training operation on an eye image sample to generate a classification parameter; and
a storage device, coupled to the training device and the computing device, and configured to store the classification parameter, such that the computing device identifies the at least one iris region in the eye image based on the classification parameter.

### Claim 8 (depends on 7)

The eye tracking apparatus as claimed in claim 7, wherein the training operation comprises model training based on multi-layer neural networks.

### Claim 9 (depends on 7)

The eye tracking apparatus as claimed in claim 7, wherein the computing device defines the at least one iris region in an enhanced eye image based on the classification parameter.

### Claim 10 (depends on 3)

The eye tracking apparatus as claimed in claim 3, wherein the computing device masks the enhanced eye image.

### Claim 11 (depends on 1)

The eye tracking apparatus as claimed in claim 1, wherein the computing device classifies the eye image into the largest iris region and a non-iris region.

### Claim 12 (depends on 11)

The eye tracking apparatus as claimed in claim 11, wherein the non-iris region comprises a first sclera region and a second sclera region assisting to position the largest iris region.

### Claim 13 (depends on 2)

The eye tracking apparatus as claimed in claim 2, wherein the gaze point is a center of the fitting pattern.

### Claim 14 (depends on 1)

The eye tracking apparatus as claimed in claim 1, further comprising:
a display device, coupled to the computing device and configured to display an output response based on the gaze point.

### Claim 15

An eye tracking method, comprising:
capturing at least one face image;
obtaining an eye image of the face image;
identifying at least one iris region in the eye image;
choosing a largest iris region from the at least one iris region and obtaining a fitting pattern of the largest iris region; and
obtaining a gaze point of the eye image based on the fitting pattern.

### Claim 16 (depends on 15)

The eye tracking method as claimed in claim 15, wherein the fitting pattern is a minimum covering circle or a minimum covering ellipse.

### Claim 17 (depends on 15)

The eye tracking method as claimed in claim 15, wherein identifying the at least one iris region in the eye image comprises:
enhancing the eye image to generate an enhanced eye image; and
identifying the at least one iris region in the enhanced eye image.

### Claim 18 (depends on 17)

The eye tracking method as claimed in claim 17, wherein the step of enhancing the eye image to generate the enhanced eye image comprises an image brightness adjustment operation, an image contrast increment operation and/or a Weickert image enhancing operation.

### Claim 19 (depends on 18)

The eye tracking method as claimed in claim 18, wherein the image brightness adjustment comprises increasing a brightness at a position with an excessively low grayscale value in the eye image.

### Claim 20 (depends on 19)

The eye tracking method as claimed in claim 19, wherein the image brightness adjustment comprises lowering a brightness at a position with an excessively high grayscale value in the eye image.

### Claim 21 (depends on 17)

The eye tracking method as claimed in claim 17, wherein identifying the at least one iris region in the eye image further comprises:
performing a training operation on an eye image sample to obtain a classification parameter;
storing the classification parameter; and
identifying the at least one iris region in the eye image based on the classification parameter.

### Claim 22 (depends on 21)

The eye tracking method as claimed in claim 21, wherein the training operation comprises model training based on multi-layer neural networks.

### Claim 23 (depends on 21)

The eye tracking method as claimed in claim 21, wherein identifying the at least one iris region in the eye image comprises: identifying the at least one iris region in an enhanced eye image based on the classification parameter.

### Claim 24 (depends on 17)

The eye tracking method as claimed in claim 17, wherein identifying the at least one iris region in the eye image further comprises:
masking the enhanced eye image.

### Claim 25 (depends on 15)

The eye tracking method as claimed in claim 15, wherein choosing the largest iris region from the at least one iris region and obtaining the fitting pattern of the largest iris region comprises:
classifying the eye image into the largest iris region and a non-iris region.

### Claim 26 (depends on 25)

The eye tracking method as claimed in claim 25, wherein the non-iris region comprises a first sclera region and a second sclera region assisting to position the largest iris region.

### Claim 27 (depends on 16)

The eye tracking method as claimed in claim 16, wherein the gaze point is a center of the fitting pattern.

### Claim 28 (depends on 15)

The eye tracking method as claimed in claim 15, further comprising:
displaying a output response based on the gaze point.
