Eye Detection Using Optical Coding to Remove Glasses Reflection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems face difficulties in detecting eyes due to light reflection from glasses, which obscures the pupil and corneal reflection points, making it challenging to monitor driver state and gaze direction.
Innovation Solution
An apparatus and method using optical coding, such as a fixed pattern filter or micro mirror array, to blur the glasses reflection point and recover the eye image by adjusting the aperture's depth of field, allowing for effective removal of noise caused by glasses reflections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If a camera flash is operated to reduce the effect of sunlight and externally incident light, then the illumination of the eye is improved, but the light is reflected from glasses and obscures the eyes
Solution Approach 1:
The camera uses periodic flashing (multiple flashes) to illuminate the eye from different angles and positions. By capturing multiple images at different time points during the periodic flashing process, the system can separate the eye reflection from the glasses reflection, as they appear at different positions in the sequence of images.
Solution Approach 2:
The system analyzes the sequence of captured images to identify the eye position and glasses reflection position. Based on this feedback, it determines the relative positions and uses this information to selectively process or remove the glasses reflection while preserving the eye image.
2Illumination intensity
If externally incident light is directly reflected from glasses, then the illumination effect is achieved, but the eyes are obscured and difficult to detect
Solution Approach 1:
By using periodic flashing to capture multiple images at different angles, the system creates a temporal sequence where the eye and glasses reflection appear at different positions. This allows precise identification and measurement of the eye location despite the presence of glasses reflection.
Solution Approach 2:
The system extracts and isolates the eye image from the composite image that contains both eye and glasses reflection. By identifying the specific position of the eye in the sequence of images and separating it from the glasses reflection components, the system achieves accurate eye detection.
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
Enables accurate eye image acquisition and recovery, overcoming the obstruction caused by light reflections from glasses, thereby improving the ability to monitor driver state and gaze direction.
Implementation Method 1
a camera configured to acquire an eye image using optical coding
Implementation Method 2
The optical coding may use a fixed pattern filter
Implementation Method 3
The optical coding may use a micro mirror array
Implementation Method 4
The optical coding may use a difference in size of a corneal reflection and a glasses reflection
Data Source
AI summary
An apparatus for detecting eyes comprises a camera configured to acquire an eye image using optical coding; and an image processor configured to perform signal processing on the eye image, wherein the signal processing comprises removing noise from the eye image caused by a glasses reflection point.