Driver Drowsiness Detection via Eye Signal Validity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current drowsiness detection systems for drivers are prone to false warnings due to measurement errors, as they do not adequately differentiate between natural eye movements and camera vibrations, and often fail to recognize microsleep as an independent danger, leading to inadequate warnings.

Innovation Solution

A method that reads in eye opening signals from both eyes, checks their validity based on predetermined criteria such as eye opening speed and head movement, and uses a combination of these signals to determine the driver's tiredness or sleep state, thereby reducing false alarms and improving detection robustness.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If eye opening signals are used to detect driver tiredness, then detection capability is improved, but false warnings increase due to measurement errors from camera vibrations and head movements

Engineering Contradiction:
Improvedetection reliabilityVSAvoideye opening measurement precision
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent segments the eye opening detection into multiple independent components: left eye signal, right eye signal, head movement signal, and camera vibration signal. Each component is processed separately and then combined through validity checking to produce the final tiredness assessment, reducing the impact of errors in any single measurement

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces intermediary signals (head movement signal and camera vibration signal) that mediate between the raw eye opening measurements and the final tiredness determination. These intermediaries help filter out false positives by identifying when apparent eye closure is actually caused by head movement or camera vibration rather than actual driver tiredness

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If drowsiness detection systems monitor driving behavior, then tiredness detection is possible, but microsleep cannot be recognized as an independent danger

Engineering Contradiction:
Improvetiredness detection accuracyVSAvoiddetection coverage for different tiredness types
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent dynamically adjusts the detection approach by monitoring multiple parameters simultaneously (eye opening degree, eye opening speed, head movement) and adapting the assessment based on the pattern of signals. This allows the system to distinguish between gradual drowsiness and sudden microsleep events, treating them as different detection scenarios with appropriate response thresholds

Inventive Principle:
Principle #15Dynamics

3Reliability

If validity criteria are applied to eye opening signals, then false warnings are reduced, but system complexity increases

Engineering Contradiction:
Improvewarning accuracyVSAvoidsignal processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent performs preliminary actions by pre-establishing validity criteria and thresholds for eye opening signals, head movement signals, and camera vibration signals. These criteria are set in advance and stored in the system, allowing rapid online validation without complex real-time calculations, thus reducing computational complexity while maintaining high warning accuracy

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3268942B1Method and device for detecting a tiredness and/or sleeping state of a driver of a vehicle
Publication Date: 2019.06.19 ROBERT BOSCH GMBH
  • EP3268942B1 patent drawingFigure 1~2
  • EP3268942B1 patent drawingFigure 3~4

AI summary

The invention relates to a method for detecting a tiredness and/or sleeping state of a driver (115) of a vehicle. The method comprises a step of reading in a first eye opening signal (210) and a second eye opening signal (220), wherein the first eye opening signal (210) represents an eye opening degree and/or a signal derived therefrom of the left eye (125a) of the driver (115) of the vehicle and the second eye opening signal (220) represents an eye opening degree and/or a signal derived therefrom of the right eye (125b) of the driver (115) of the vehicle. The method also comprises a step of detecting a validity of the first and/or second eye opening signal signal (210, 220) in order to recognize the first eye opening signal (245) as valid if the first eye opening signal (210) satisfies a first criterion and/or in order to recognize the second eye opening signal (250) as valid if the second eye opening signal (220) satisfies a second criterion. Finally, the method comprises a step of determining the tiredness and/or sleeping state of the driver (115) of the vehicle (100) by using the first eye opening signal (245) recognized as valid and/or the second eye opening signal (250) recognized as valid.