Infrared Driver Monitoring Under Sunlight Flare Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving assistance systems inaccurately determine driver anomalies due to flared highlights caused by sunlight in infrared images, leading to unnecessary warnings when the driver is actually driving normally.
Innovation Solution
A driving assistance apparatus that acquires infrared images, determines the occurrence of flared highlights based on pixel values, and adjusts the warning time accordingly to avoid false alarms by extending the waiting period when flared highlights are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system determines driver anomaly based on infrared images without considering flared highlights, then the driver anomaly detection is simple and fast, but false warnings occur when sunlight hits the driver's seat
Solution Approach 1:
The system performs preliminary detection of flared highlights by checking pixel values in sunlight-hit areas before making the final driver anomaly determination. This preliminary action prevents false warnings by identifying and excluding images with flared highlights from the anomaly detection process
Solution Approach 2:
The system introduces an intermediary determination process that checks for flared highlights as a intermediate step between acquiring infrared images and determining driver anomalies. This intermediary check acts as a filter to prevent false warnings while maintaining the overall detection framework
2Speed
If the system gives immediate warnings when driver recognition fails, then the response time is fast, but unnecessary warnings are given due to flared highlights
Solution Approach 1:
The system performs preliminary checking for flared highlights before issuing warnings. By checking pixel values in the determination portion before the warning decision, the system filters out false cases while maintaining fast response for genuine anomalies
Solution Approach 2:
The system uses feedback from the flared highlight detection process to adjust the warning issuance. When flared highlights are detected, the system suppresses false warnings; when no flared highlights are present, the system proceeds with normal warning issuance, creating a feedback-based warning control mechanism
3Reliability
If the system checks pixel values to detect flared highlights, then false warnings are reduced, but the processing time and computational load increase
Solution Approach 1:
The system applies pixel value checking only to specific determination portions that are hit by sunlight, rather than processing the entire infrared image. This localized approach reduces computational load while effectively detecting flared highlights in the relevant areas
Solution Approach 2:
The system changes the detection parameter from full-image processing to specific pixel value threshold checking in sunlight-hit areas. By monitoring pixel values against predetermined thresholds only in affected regions, the system achieves reliable flared highlight detection with minimal processing overhead
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Reduces false warnings by accurately distinguishing between normal driving postures and anomalies, ensuring timely alerts only when necessary.
Implementation Method 1
an infrared light that irradiates a first area including a driver's seat where a driver of a vehicle is seated with an infrared ray
Implementation Method 2
an image-capturing apparatus that captures an image of the first area
Data Source
AI summary
A driving assistance apparatus has: an acquiring section that acquires an infrared image capturing an image capturing area including a driver's seat of a vehicle; a determining section that determines whether a flared highlight has occurred in the infrared image in a case where it is not possible to recognize the driver in the infrared image, on the basis of whether a pixel value of a pixel of a portion capturing an image of an area which is in the image capturing area in the infrared image and is hit by sunlight is equal to or greater than a predetermined value; and a setting section that makes waiting time longer than reference time, the waiting time being a time point at which a warning is given in a case where it is determined that the flared highlight has occurred in the infrared image.