Blur Estimation in Eye Images Using Iris-Pupil Edge Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric systems face challenges in acquiring sharply-focused images of moving subjects due to image blur, which is ill-posed in the general case and requires estimates of blur magnitude and shape, particularly for iris-based biometrics where motion and optical blur are prevalent.
Innovation Solution
A feature-based method and system for blur estimation in eye images utilizing the edge between the iris and pupil regions, shutter motion patterns, and additional image information from a burst of images or video stream to predict eye motion, allowing for de-blurred images suitable for biometric identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If blur estimation is performed in the general case, then the problem is ill-posed and cannot be solved, but in eye images additional information is available to make it solvable
Solution Approach 1:
The patent segments the eye image into distinct anatomical regions (iris, pupil, sclera, limbus) and uses the specific geometric relationships between these regions to constrain the blur estimation problem. By dividing the image into meaningful segments with known spatial relationships, the ill-posed general problem becomes solvable for eye images specifically.
Solution Approach 2:
The patent changes the parameters used for blur estimation by incorporating eye-specific anatomical constraints (fixed radial geometry of iris-pupil boundary, known color transitions at limbus) rather than using general image features. This parameter transformation makes the previously ill-posed problem solvable in the eye image domain.
2Reliability
If deblurring is performed to improve image quality for biometric identification, then image quality improves, but the process requires accurate estimates of blur magnitude, motion direction, and shutter pattern which are difficult to obtain
Solution Approach 1:
The patent uses feedback from detected eye features (iris-pupil boundary, limbus location) to iteratively refine blur parameter estimates. The detected anatomical features provide feedback constraints that guide the deblurring process and improve the precision of blur magnitude and direction estimates.
Solution Approach 2:
The patent transforms the difficult problem of estimating general blur parameters into estimating eye-specific parameters (radial blur direction from pupil center, magnitude from iris edge sharpness) which can be more accurately determined from the image data.
3Ease of operation
If standard image capture settings are used, then capture is straightforward, but image blur occurs due to subject motion or optical defocus
Solution Approach 1:
The system performs self-correction by automatically detecting eye features and estimating blur parameters from the captured image itself, then applying deblurring without requiring external intervention or specialized capture equipment. The image captures for itself the information needed to correct its own blur.
Solution Approach 2:
The patent replaces mechanical focus adjustment and motion control systems with computational deblurring based on eye feature analysis. Instead of using mechanical means to prevent blur (stabilizers, focus mechanisms), it uses image processing substituting mechanical correction with computational correction.
Data Source
AI summary
A feature-based method and system for blur estimation in eye images. A blur estimation can be performed from eye/iris images in order to produce de-blurred images that are more useful for biometric identification. The eye/iris region, in particular the edge between the iris and pupil regions, can be utilized. The pattern of shutter motion or a characterization of the optical system can be utilized. By capturing a burst of images, or a video stream, one can use eye position in the images before and after a given capture to predict the motion of the eye within that capture. Because the before/after image frames need only contain the information necessary to locate the eye, and need not contain sufficient information to perform matching, the capture of these images can be accomplished with a wider range of settings.