Iris Recognition Excluding Eyelash Occlusions via Threshold Brightness
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current iris recognition technologies face challenges in accurately distinguishing iris features from images that include occlusions, such as eyelashes, which can lead to reduced recognition accuracy and reliability.
Innovation Solution
The method involves obtaining an iris image, extracting a reference area by dividing it into candidate and reference regions based on axis orientations, determining a threshold brightness from the reference area's histogram, and excluding pixels below this threshold to enhance feature extraction and recognition accuracy, thereby improving iris recognition accuracy even with irregular occlusions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional iris recognition methods are used without excluding occluding elements, then the recognition process is simpler, but the recognition accuracy deteriorates due to interference from eyelashes and other occlusions
Solution Approach 1:
The iris image is divided into multiple candidate areas based on axis orientations (horizontal, vertical, diagonal axes). Each candidate area is processed separately to identify occluding elements, allowing the system to segment the problem of occlusion removal into manageable regions rather than processing the entire image uniformly.
Solution Approach 2:
The system changes the brightness parameter threshold dynamically by determining a threshold brightness value from the reference area and using it to identify and exclude target pixels (occlusions) in candidate areas. This parameter-based approach automatically adapts to different imaging conditions without requiring manual intervention.
2Reliability
If a simple threshold-based method is used to exclude occlusions, then the processing speed is faster, but the reliability deteriorates when occlusion patterns are complex or irregular
Solution Approach 1:
The system performs preliminary actions by first determining the threshold brightness from a reference area before applying it to exclude occlusions in candidate areas. This preliminary threshold determination ensures that the subsequent occlusion exclusion process is both reliable and efficient, as the threshold is pre-calculated based on the specific image characteristics.
Solution Approach 2:
The threshold brightness value acts as an intermediary parameter that mediates between the reference area characteristics and the occlusion exclusion process. This intermediary enables the system to reliably identify occlusions across different candidate areas while maintaining processing efficiency through a unified threshold-based approach.
3Measurement precision
If the entire iris image is processed for feature extraction, then more iris features are captured, but the processing time increases and occluding elements interfere with accurate feature extraction
Solution Approach 1:
The system extracts and excludes target pixels (occluding elements) from candidate areas based on the threshold brightness determined from the reference area. By removing these interfering elements before feature extraction, the system improves feature extraction accuracy while reducing processing time compared to processing the entire unfiltered image.
Solution Approach 2:
The system applies different processing qualities to different regions: the reference area is used to determine threshold characteristics, while candidate areas have occlusions excluded based on this threshold. This local quality approach ensures that feature extraction is performed on cleaned, high-quality data from candidate areas without unnecessarily processing occluded regions.
Data Source
AI summary
A device and method of recognizing an iris is provided. An iris recognition device may determine a threshold brightness using a histogram of a reference area from an iris image and exclude a target pixel from iris recognition in a candidate area based on the determined threshold brightness.