Driver Gaze Correlation for In-Vehicle Mobile Use Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driver monitoring systems face challenges in accurately detecting driver distraction caused by mobile devices, particularly distinguishing between driver and passenger use, and effectively addressing the safety hazard of attention diversion from driving tasks.
Innovation Solution
A method and system that utilize camera-based image processing to determine visual attention of the driver by analyzing head and eye movements, correlating gaze direction with mobile device use events, and classifying regions of interest to identify when a driver is using a mobile device, with thresholds for determining distraction based on gaze time and vehicle velocity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If antenna-based localization or computer-vision approaches are used to detect mobile devices, then device location can be determined, but false detections occur when devices are within passenger reach zones
Solution Approach 1:
The system segments the detection problem into two independent components: (1) mobile device location detection using antennas or computer vision, and (2) driver visual attention monitoring using eye-tracking cameras. By separating these functions, the system avoids false positives from passenger device usage while maintaining accurate driver distraction detection through gaze analysis.
Solution Approach 2:
The system introduces an intermediary measurement - driver gaze direction and visual attention - to mediate between device location and distraction determination. Instead of directly inferring distraction from device presence, the system uses gaze data as an intermediary to confirm whether the driver is actually attending to the device, thereby resolving false detection issues.
2Measurement precision
If the system monitors driver gaze and correlates with device use events, then driver distraction can be accurately detected, but system complexity increases
Solution Approach 1:
The system employs a multi-functional integrated camera that simultaneously performs multiple tasks: capturing driver facial images for gaze analysis, detecting mobile device presence through computer vision, and monitoring overall driver behavior. This universal approach reduces the number of separate components needed while maintaining high detection accuracy.
Solution Approach 2:
The system merges previously separate functions - device detection, gaze tracking, and distraction determination - into a unified monitoring framework. By combining these functions into a single integrated system that processes multiple data streams simultaneously, the complexity of managing separate systems is reduced while maintaining comprehensive monitoring capabilities.
3Adaptability or versatility
If the system uses multiple sensing methods (radio, ultra-sonic, visual) to locate devices, then detection coverage is improved, but false alarms increase due to overlapping driver and passenger zones
Solution Approach 1:
Instead of trying to prove that a device is being used by the driver through multiple sensing methods, the system inverts the approach: it assumes distraction unless the driver's gaze clearly indicates attention to the device. This inversion reduces false alarms by requiring positive evidence of driver attention rather than attempting to exclude all other possibilities.
Data Source
AI summary
Described herein is a method (1000) of detecting mobile device (400) use of a driver (102) of a vehicle (104). The method (1000) comprises the step (1001) of receiving a sequence of images of at least the driver's head captured from a camera (106, 420). At step (1002), the sequence of images are processed to determine visual attention of the driver (102) based on detected head and/or eye movements of the driver (102) over a period of time. At step (1003), mobile device use events are detected within the period of time in which a user interacts with the mobile device (400) that is located within the vehicle (104). At step (1004), a temporal correlation of the visual attention of the driver (102) with the mobile device use events is determined over the period of time. At step (1005), a determination is made that the driver (102) is using the mobile device (400) if the determined temporal correlation is greater than a threshold correlation coefficient.