Driver Device-Use Detection from Face Orientation and Hidden Pupils
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The use of handheld electronic devices by drivers poses a significant distraction, endangering both the driver and other road users, necessitating a reliable method to detect such usage.
Innovation Solution
A method involving image analysis to determine the orientation of a driver's face and the presence of pupils in captured images, combined with time and frequency thresholds, to infer handheld device usage, which is enhanced by considering vehicle autonomy and movement, and confirmed by steering corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single condition (face orientation or pupil presence) is used to detect handheld device usage, then the detection process is simpler, but the reliability of detection decreases
Solution Approach 1:
The patent combines multiple detection conditions (face orientation analysis and pupil presence detection) into a unified detection system. Both conditions must be satisfied simultaneously to infer handheld device usage, merging separate detection functions into a coordinated system that improves reliability while maintaining manageable complexity through integrated processing.
Solution Approach 2:
The system performs preliminary analysis of face orientation and pupil presence before making a final detection decision. By pre-processing and evaluating multiple indicators in advance, the system builds a comprehensive assessment that enhances detection reliability while organizing complexity into structured preliminary steps.
2Measurement precision
If the detection system monitors multiple conditions simultaneously, then the detection accuracy improves, but the processing time increases
Solution Approach 1:
The detection system operates by periodically capturing images and evaluating multiple conditions at discrete time intervals rather than continuous monitoring. This periodic approach maintains high detection accuracy through multiple condition checks while controlling processing time by evaluating all conditions at scheduled intervals rather than simultaneously in real-time.
3Reliability
If the detection method considers various face orientations and pupil positions, then the false detection rate decreases, but the complexity of the detection algorithm increases
Solution Approach 1:
The detection algorithm segments the analysis into distinct functional modules: face detection, orientation analysis, pupil presence detection, and final inference. Each module handles a specific aspect of the detection task, reducing overall algorithm complexity while maintaining high reliability through comprehensive coverage of different face orientations and pupil positions within each segment.
Data Source
Figure 1~2
AI summary
The disclosure relates to a method for detecting that a driver (17) of a vehicle (16) is using a handheld electronic device (20). The method comprises capturing or receiving at least one image (I) of the vehicle (16) showing at least a partial frontal view of the vehicle (16) including the driver (17) and determining a face orientation (G) of the driver (17) based on the at least one captured image (I). Moreover, the method comprises determining whether at least one pupil (38) of the driver (17) is shown in the at least one captured image (I). Subsequently, it is inferred that the driver (17) is using the handheld electronic device (20) if the face orientation (G) is downwards and no pupil (38) of the driver (17) is shown or if the face orientation (G) is straight ahead and no pupil (38) of the driver (17) is shown. Furthermore, a method for controlling a leading vehicle (10) travelling in front of a following vehicle (16) is described. Additionally, a data processing apparatus (24), a computer program (32), and a computer-readable storage medium (30) are presented. Moreover, a system (18) for detecting that a driver (17) of a following vehicle (16) is using a handheld electronic device (20) is explained.