Primary Preview Region Gaze Detection for Driver Distraction Alerts

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distracted driving leads to significant accidents and loss of life, with existing systems failing to effectively alert drivers to critical road conditions due to insufficient attention to primary preview regions.

Innovation Solution

A system that uses gaze detection and machine learning to identify primary preview regions (PPRs) and generate alerts when a driver's attention falls below a threshold, employing cameras and machine learning to determine PPRs and adjust attention levels, with alerts via HUD, haptic feedback, or automated vehicle control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If gaze detection and machine learning are used to identify primary preview regions and monitor driver attention, then driver safety is improved by alerting distracted drivers to critical road conditions, but device complexity increases due to the need for cameras, machine learning algorithms, and multiple alert systems

Engineering Contradiction:
Improvedriver safetyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the road environment into multiple primary preview regions (PPRs) based on hazard levels and driver attention needs. Each PPR is independently monitored and can trigger separate alert conditions, allowing the complex monitoring task to be divided into manageable regional segments rather than treating the entire road scene as a single unit

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an attention level as an intermediary metric that mediates between raw gaze detection data and final alert generation. The system calculates attention levels for each PPR based on gaze duration and proximity, using this intermediate measure to determine when alerts should be triggered, thereby simplifying the decision-making process between complex sensor data and binary alert states

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the system monitors multiple primary preview regions with different attention levels, then detection precision of driver distraction is improved, but computational load and processing time increase

Engineering Contradiction:
Improvedistraction detection precisionVSAvoidcomputational load
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The system applies local quality by assigning different attention levels to different PPRs based on their hazard characteristics. High-hazard regions require higher attention thresholds and more intensive monitoring, while low-hazard regions use lower thresholds. This localized differentiation improves detection precision for critical areas without uniformly increasing computational load across the entire scene

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system implements partial monitoring by focusing computational resources on a limited number of primary preview regions rather than analyzing the entire road scene. By identifying and monitoring only the most relevant PPRs where hazards are likely to occur, the system achieves sufficient distraction detection precision while avoiding the excessive computational burden of comprehensive scene analysis

Inventive Principle:
Principle #16Partial or excessive action

3Loss of information

If alerts are generated based on attention level thresholds for each PPR, then the system's ability to provide targeted safety warnings is improved, but the quantity of information processed and alert signals increases

Engineering Contradiction:
Improveinformation retentionVSAvoidalert signals
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The system employs dynamic alert generation where the decision to issue alerts depends on the current attention level relative to dynamically adjusted thresholds for each PPR. Attention levels are continuously updated based on real-time gaze data, and thresholds are adjusted according to hazard characteristics, creating a dynamic monitoring system that adapts to changing driving conditions rather than using static, fixed criteria

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3735365B2Primary preview region and gaze based driver distraction detection
Publication Date: 2025.08.20 HUAWEI TECH CO LTD
  • EP3735365B2 patent drawingFigure 1
  • EP3735365B2 patent drawingFigure 2
  • EP3735365B2 patent drawingFigure 3

AI summary

A computer-implemented method of detecting distracted driving comprises: determining, by one or more processors, a primary preview region (PPR) in a representation of an environment; determining, by the one or more processors, a gaze point for a driver based on a sequence of images of the driver; determining, by the one or more processors, that the gaze point is outside of the PPR; based on the determined gaze point being outside of the PPR, decreasing, by the one or more processors, an attention level for the PPR; based on the attention level for the PPR, generating, by the one or more processors, an alert.