Driver State Determination Using Face Feature Frequency Distribution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing driver state determination systems require a preset operation from the driver to set the allowable distraction determination range, which is inconvenient and may not account for individual differences in visual field range and posture.

Innovation Solution

A driving state determination device that detects face feature points from a driver's face image, creates a frequency distribution of face information, calculates a mode value, and uses this to determine a steady state reference value for determining the driving state without requiring a preset operation from the driver.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a preset operation is required from the driver to set the allowable distraction determination range, then the system can establish personalized parameters for each driver, but the ease of operation deteriorates due to the additional manual setup requirement

Engineering Contradiction:

Improveaccuracy of driving state determinationVSAvoidconvenience of system setup

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The system automatically captures face images during normal driving and performs self-calibration by detecting face feature points and calculating reference values without requiring any manual operation from the driver. The calibration process serves itself by using the driver's own facial characteristics to establish personalized parameters.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs calibration actions in advance during normal driving operations before actual distraction detection is needed. By establishing the reference values and determining the driver's visual field range during regular use, the system prepares personalized parameters proactively rather than requiring separate setup procedures.

Inventive Principle:

Principle #10Preliminary action

2Ease of operation

If the visual field range is set using a fixed normal range for all drivers, then the ease of operation is maintained, but the measurement precision deteriorates due to individual differences in visual field range and posture

Engineering Contradiction:

Improvesimplicity of system configurationVSAvoidaccuracy of distraction detection

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system transitions from a uniform normal range applicable to all drivers to personalized reference values specific to each driver's facial characteristics and visual field range. By detecting individual face feature points and calculating unique reference values, the system adapts the determination criteria to match each driver's local characteristics.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts the reference values and visual field range parameters based on detected face information and frequency distribution analysis. Instead of using fixed parameters, the system modifies these parameters to reflect individual driver characteristics, thereby improving measurement precision while maintaining operational simplicity.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If manual preset operation is required for setting calibration parameters, then the device complexity remains low, but the productivity deteriorates due to the time required for manual configuration

Engineering Contradiction:

Improvesimplicity of system architectureVSAvoidspeed of system deployment

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The system automatically performs calibration by capturing face images and computing reference values without requiring manual intervention. This self-calibration capability eliminates the time-consuming manual configuration step while maintaining relatively simple device architecture that leverages existing face detection and image processing functions.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces the mechanical interaction of manual preset operations with automated computational processes. By using frequency distribution analysis and mode value calculation algorithms, the system substitutes manual configuration actions with automatic data processing, thereby improving productivity without significantly increasing device complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10949690B2Driving state determination device, determination device, and driving state determination method

Publication Date: 2021.03.16 MITSUBISHI ELECTRIC MOBILITY CORP

AI summary

There are included: a frequency distribution creating unit for referring to the face information detected by the face information detecting unit, and when a preset change occurs in the driver's face, creating a frequency distribution of face information in a preset time section from the face information detected by a face information detecting unit; a mode value calculating unit for calculating a mode value of the face information from the frequency distribution created; and a reference value calculating unit for calculating a reference value indicating a steady state of the driver from the mode value of the face information.