Driver Observation Data Processing for Privacy-Safe Mode Transitions

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing automatic driving technologies face challenges in ensuring driver safety during mode transitions from automatic to manual driving, particularly when the driver's arousal level is low, and there is a need to record observation information without violating personal information protection regulations.

Innovation Solution

An information processing device and system that acquires driver observation information, divides it into conversion unnecessary and necessary data, applies abstraction or encryption processing to the necessary data, and records the converted data while maintaining confidentiality, allowing for safe mode transitions and accident analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If driver observation information is recorded to ensure safety during mode transitions, then driver safety is improved, but personal information protection requirements are violated

Engineering Contradiction:

Improvedriver safetyVSAvoidpersonal information protection violation

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system extracts only the necessary safety-critical information (arousal level, gaze direction, head position) from the captured driver image, separating it from identifiable personal information. This extracted data is then used for safety monitoring while the original identifiable image data is not stored, thus ensuring both safety monitoring and personal information protection

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system introduces an intermediary processing stage that captures the driver image, extracts safety-relevant features (arousal level, gaze direction, head position), and then discards the original identifiable image. This intermediary process acts as a mediator between safety monitoring requirements and personal information protection, transforming raw image data into anonymized safety metrics

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If complete driver image data is stored for accident analysis, then evidence quality is improved, but data privacy requirements are violated

Engineering Contradiction:

Improveevidence qualityVSAvoiddata privacy violation

Core Design Contradiction:

Loss of informationVSObject-affected harmful factors

Solution Approach 1:

The system extracts only the essential safety-critical features (arousal level, gaze direction, head position) from the driver image, separating them from identifiable personal information. This extracted data provides sufficient evidence for accident analysis without containing privacy-violating personal identifiers

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of the conventional approach of storing complete images and then filtering, the system inverts the process by first extracting only the necessary safety features and immediately discarding the original identifiable image data. This inversion ensures that only anonymized safety-critical information is retained for evidence purposes

Inventive Principle:

Principle #13The other way round (Inversion)

3Reliability

If driver monitoring is implemented during automatic driving, then mode transition safety is improved, but system complexity increases

Engineering Contradiction:

Improvemode transition safetyVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system uses a single camera to perform multiple functions: capturing driver images for arousal level detection, gaze direction analysis, and head position monitoring. This multi-functional use of the camera reduces overall system complexity compared to using separate sensors for each monitoring function

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system processes driver monitoring data entirely autonomously without requiring external intervention. The camera captures images, the processing unit analyzes arousal level and head position, and the system automatically determines readiness for mode transition, making the monitoring system self-sufficient and reducing operational complexity

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250304121A1Information processing device, mobile device, information processing system, and method

Publication Date: 2025.10.02 SONY SEMICON SOLUTIONS CORP

AI summary

Provided is a driver information acquisition unit that acquires the observation information of the driver of the vehicle and a data processing unit that inputs the observation information and executes data processing are included. The data processing unit divides the observation information into conversion unnecessary data and conversion necessary data, executes abstraction processing or encryption processing for the conversion necessary data, and stores conversion data such as abstraction data or encryption data in a storage unit. The data processing unit executes the abstraction processing or the encryption processing for the individually-identifiable data included in the observation information or the data for which recording processing is not permitted in the personal information protection regulation as the conversion necessary data.