Self-Calibrating Driver Head Orientation Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing driver monitoring systems face challenges in accurately detecting driver inattention due to difficulties in calibrating imaging sensors, particularly in vehicles with larger cab spaces, leading to false detections and inability to adapt to different drivers or changes in driver posture.

Innovation Solution

The system uses imaging sensors to capture and process images of a driver's head to identify the natural resting orientation, averaging multiple measurements to determine a driver's average head orientation, which is then used to detect deviations indicative of inattention, allowing for real-time and dynamic calibration without requiring manual installer input.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If manual calibration by installer is used, then initial setup is simple, but accuracy deteriorates in large cab spaces and cannot adapt to different drivers

Engineering Contradiction:

Improveease of calibrationVSAvoidhead orientation detection accuracy

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system performs automatic calibration by capturing images of the driver's head and computing orientation parameters without requiring manual installer input. The processing device automatically identifies head features, calculates orientation angles, and stores calibration data, enabling the system to self-calibrate for each driver individually.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system dynamically adjusts calibration parameters by capturing multiple images and computing average head orientation parameters. It adapts to different drivers by changing the stored baseline orientation parameters based on actual driver measurements rather than using fixed installer-defined values.

Inventive Principle:

Principle #35Parameter changes

2Device complexity

If fixed calibration is used, then system is simple, but adaptability deteriorates for different drivers and changing postures

Engineering Contradiction:

Improvecalibration system complexityVSAvoidadaptation to different drivers

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system transitions from static fixed calibration to dynamic adaptive calibration. It continuously monitors head orientation and recalibrates by computing averages from multiple captured images, allowing the calibration parameters to adapt dynamically to different drivers and changing conditions while maintaining reasonable system complexity.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The calibration system serves multiple functions: it calibrates for different drivers individually, adapts to changing driver postures, and provides continuous self-adjustment. The same processing device handles both initial calibration and ongoing adaptation, making the system universally applicable to various drivers and conditions.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Loss of time

If single image calibration is used, then processing is fast, but precision deteriorates due to posture variations

Engineering Contradiction:

Improvecalibration timeVSAvoidhead orientation accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system performs preliminary calibration actions by capturing multiple images before determining the final head orientation parameters. It pre-processes multiple frames and computes averages in advance, establishing an accurate baseline that compensates for posture variations before actual monitoring begins.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The calibration process continues by capturing multiple images and continuously computing average orientation parameters. Rather than using a single snapshot, the system maintains continuous image capture and processing to establish robust calibration data that accounts for natural head movements and posture variations.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS11600083B1Highly-accurate and self-adjusting imaging sensor auto-calibration for in-vehicle driver monitoring system or other system

Publication Date: 2023.03.07 OMNITRACS LLC

AI summary

A method includes obtaining multiple images of a driver of a vehicle using an imaging sensor associated with the vehicle, where the images of the driver capture the driver's head. The method also includes identifying, in each of at least some of the images, an orientation of the driver's head in the image. The method further includes identifying an average orientation of the driver's head based on at least some of the identified orientations of the driver's head. In addition, the method includes determining whether the driver is inattentive based on the average orientation of the driver's head. Identifying, in each of at least some of the images, the orientation of the driver's head in the image may include identifying a pitch angle and a yaw angle of the driver's head in the image.