Distraction detection method in a driving monitoring system

JP2026508812APending Publication Date: 2026-03-13ZYLON DAY OH OH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-01-05
Publication Date
2026-03-13

AI Technical Summary

Technical Problem

Existing driver distraction detection systems fail to account for the interaction between the driver's gaze zones and the influence of the driver's expected future gaze choices, relying solely on current gaze zone scoring without considering the relevance in the driving situation and recorded gaze history.

Method used

A scoring method that incorporates Potential Usefulness Level (PUL) values for all zones, dynamically updating these values based on current and future gaze interactions, and using a four-level scoring system to issue warnings, including visual, audio, and automatic brake signals.

Benefits of technology

Improves the detection of secondary tasks by considering the interrelationships between gaze zones and their relevance in the driving context, providing more accurate and timely warnings to prevent distractions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026508812000001_ABST
    Figure 2026508812000001_ABST
Patent Text Reader

Abstract

This disclosure provides a method for operating a distraction detection module (40) used in a driving monitoring system, which provides driver gaze zone estimation (30) data captured for each frame predefined for a particular vehicle, and vehicle information (15) captured from the vehicle, such as a controller area network (CAN) bus (14). The method performs scoring of multiple time windows using values ​​in the range of 0 to 1 based on the potential utility level (PUL) of the gaze zone recognized for each frame, and further creates a data vector to extract possible driver secondary activities. Finally, the instantaneous distraction level of the driver is estimated as a discrete output {0, 1, 2, 3} using a specialized method. Here, the gaze zones are reclassified for each captured frame, and the PUL table for those zones is dynamically updated.
Need to check novelty before this filing date? Find Prior Art