# US 2025/0148809 A1 · Driver abnormality sign detection device

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

- Number: US 2025/0148809 A1
- Document ID: US20250148809A1
- Status: Published application
- Kind: A1
- Application number: 18/940,893
- Filed: 2024-11-08
- Published: 2025-05-08
- Earliest claimed: 2023-11-08
- Assignees: Mazda Motor Corporation
- Applicants: Mazda Motor Corporation
- Inventors: Junichiro KUWAHARA; Fumito YAMANO
- Web page: https://patentgrep.com/patent/US20250148809A1

## CPC

- G06V 20/597: {Recognising the driver's state or behaviour, e.g. attention or drowsiness}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A driver abnormality sign detection device includes a control circuit configured to detect an abnormality sign of a driver based on visual line information acquired from a visual line detector and on image data from a vehicle-outside camera. The control circuit is configured to acquire reference saccades of the driver, calculate dispersion of a section including a peak of saliency in a visual field of the driver based on the image data, calculate predicted saccades by correcting the reference saccades with correction values acquired based on the dispersion of the section including the peak of the saliency, and detect presence or absence of the abnormality sign of the driver based on a visual line abnormality degree representing an extent that measured saccades diverge from the predicted saccades.

## Claims

### Claim 1

A driver abnormality sign detection device detecting an abnormality sign of a driver driving a vehicle, the driver abnormality sign detection device comprising:
a control circuit configured to detect presence or absence of the abnormality sign of the driver based on visual line information acquired from a visual line detector and on image data acquired from a vehicle-outside camera, wherein
the control circuit is configured to
acquire a reference saccade frequency and/or a reference saccade amplitude of the driver,
calculate dispersion of a section including a peak of saliency in a saliency distribution in a visual field of the driver based on the image data,
acquire correction values for correcting the reference saccade frequency and/or the reference saccade amplitude based on the dispersion of the section including the peak of the saliency,
calculate a predicted saccade frequency and/or a predicted saccade amplitude by correcting the reference saccade frequency and/or the reference saccade amplitude with the correction values,
calculate a visual line abnormality degree which represents an extent that a measured saccade frequency and/or a measured saccade amplitude based on the visual line information diverge from the predicted saccade frequency and/or the predicted saccade amplitude, and
detect, based on the visual line abnormality degree, presence or absence of the abnormality sign of the driver.

### Claim 2 (depends on 1)

The driver abnormality sign detection device according to claim 1, wherein
the correction values are set to correct the reference saccade frequency and/or the reference saccade amplitude further in a direction in which the reference saccade frequency and/or the reference saccade amplitude increase as the dispersion of the section including the peak of the saliency becomes larger.

### Claim 3 (depends on 2)

The driver abnormality sign detection device according to claim 2, wherein
the correction value is set to correct the reference saccade frequency further in a direction in which the reference saccade frequency increases as the dispersion of the section including the peak of the saliency becomes larger, and
the control circuit is configured to
acquire the reference saccade frequency and
calculate the predicted saccade frequency by correcting the reference saccade frequency with the correction value.

### Claim 4 (depends on 3)

The driver abnormality sign detection device according to claim 3, wherein
the control circuit is configured to
acquire both the reference saccade frequency and/or the reference saccade amplitude of the driver,
calculate both the predicted saccade frequency and the predicted saccade amplitude by correcting the reference saccade frequency and the reference saccade amplitude with the correction values,
accumulate two-dimensional data in which a difference between the measured saccade frequency and the predicted saccade frequency is set as a first variable and a difference between the measured saccade amplitude and the predicted saccade amplitude is set as a second variable, and
calculate the visual line abnormality degree based on a Mahalanobis distance between a newest data point of the two-dimensional data and a centroid of a set of the accumulated two-dimensional data.

### Claim 5 (depends on 2)

The driver abnormality sign detection device according to claim 2, wherein
the control circuit is configured to
acquire predetermined reference values of both the saccade frequency and the saccade amplitude of the driver,
calculate both the predicted saccade frequency and the predicted saccade amplitude by correcting the reference saccade frequency and the reference saccade amplitude with the correction values,
accumulate two-dimensional data in which a difference between the measured saccade frequency and the predicted saccade frequency is set as a first variable and a difference between the measured saccade amplitude and the predicted saccade amplitude is set as a second variable, and
calculate the visual line abnormality degree based on a Mahalanobis distance between a newest data point of the two-dimensional data and a centroid of a set of the accumulated two-dimensional data.

### Claim 6 (depends on 1)

The driver abnormality sign detection device according to claim 1, wherein
the control circuit is configured to
acquire both the reference saccade frequency and reference saccade amplitude of the driver,
calculate both the predicted saccade frequency and the predicted saccade amplitude by correcting the reference saccade frequency and the reference saccade amplitude with the correction values,
accumulate two-dimensional data in which a difference between the measured saccade frequency and the predicted saccade frequency is set as a first variable and a difference between the measured saccade amplitude and the predicted saccade amplitude is set as a second variable, and
calculate the visual line abnormality degree based on a Mahalanobis distance between a newest data point of the two-dimensional data and a centroid of a set of the accumulated two-dimensional data.

### Claim 7

A non-transitory computer readable storage device having computer readable instructions that when executed by circuitry cause the circuitry to:
acquire a reference saccade frequency and/or a reference saccade amplitude of the driver,
calculate dispersion of a section including a peak of saliency in a saliency distribution in a visual field of the driver based on image data from a vehicle-outside camera,
acquire correction values for correcting the reference saccade frequency and/or the reference saccade amplitude based on the dispersion of the section including the peak of the saliency,
calculate a predicted saccade frequency and/or a predicted saccade amplitude by correcting the reference saccade frequency and/or the reference saccade amplitude with the correction values,
calculate a visual line abnormality degree which represents an extent that a measured saccade frequency and/or a measured saccade amplitude based on visual line information from visual line information detector diverge from the predicted saccade frequency and/or the predicted saccade amplitude, and
detect, based on the visual line abnormality degree, presence or absence of the abnormality sign of the driver.
