Glint Detection in Iris Images for Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In iris recognition systems, glints created by the reflection of light during pupil center detection distort iris data, making it insufficient for authentication, leading to errors and reduced recognition rates.
Innovation Solution
A method to detect glints in iris images by identifying rows and columns with consecutive pixels above a threshold brightness value, selecting consecutive segments, and determining the shape of these pixels to exclude glint areas, ensuring only sufficient iris data is used for authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If light is irradiated from the illumination unit to induce pupil size change, then the pupil center can be found, but a glint is created that distorts iris data
Solution Approach 1:
The patent applies preliminary action by detecting and identifying glint regions in the iris image before the actual iris recognition process. The system scans the image to locate bright regions (glints) and marks them for exclusion, ensuring that only valid iris data is used for authentication. This preliminary detection prevents glint-induced distortion from affecting the iris data extraction and matching processes.
2Quantity of substance
If multiple iris images are captured for authentication, then sufficient iris data can be obtained, but images with large glints cannot be used
Solution Approach 1:
The patent applies parameter changes by using brightness value thresholds to identify and exclude glint-affected regions. The system sets a predetermined brightness threshold to distinguish glint pixels from valid iris pixels, and adjusts the data collection parameters to ensure sufficient valid iris data is captured for authentication. This parameter-based filtering ensures that only images with adequate valid iris data are used for authentication.
3Reliability
If glint detection is performed to exclude distorted images, then recognition rate increases, but the detection process adds complexity
Solution Approach 1:
The patent applies segmentation by dividing the iris image into individual pixels and analyzing their brightness values independently. The system segments the image data row by row and column by column, identifying glint regions through pixel-level brightness comparison. This segmentation approach allows for precise glint detection without requiring complex image processing algorithms, maintaining system efficiency while improving recognition reliability.
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
Accurately detects glints, increases recognition rates, and decreases error rates by excluding distorted iris images, thereby improving the reliability of user authentication.
Implementation Method 1
the light irradiated from the illumination unit is reflected off the pupil such that a 'glint' is inevitably created
Data Source
AI summary
Receiving an iris image; detecting, among rows of the iris image, rows in each of which a number of consecutive pixels each having a brightness value above a first threshold value is larger than a second threshold value; detecting, among columns of the iris image, columns in each of which a number of consecutive pixels each having a brightness value above the first threshold value is larger than the second threshold value; selecting, among the detected rows, consecutive rows in a vertical direction whose number is larger than a third threshold; selecting, among the detected columns, consecutive columns in a horizontal direction whose number is larger than the third threshold and determining a set of the pixels as a glint if the set of pixels included in the selected rows and the selected columns and each having the brightness value above the first threshold has a predetermined shape.