Iris Sub-Region Authentication for Low-Cost Recognition Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ensemble learning methods for iris recognition face challenges in achieving high recognition performance without requiring expensive equipment and strict image capture conditions, and traditional ensemble learning methods involve high learning costs and require trial and error.
Innovation Solution
The information processing device selects multiple different sub-regions of the iris based on eye features, calculates features for each sub-region, and authenticates individuals based on the similarity of these sub-regions, using a combination of region selection, size conversion, and similarity calculation to enhance recognition performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional ensemble learning methods are used for iris recognition, then recognition performance can be improved, but learning costs increase and require trial and error
Solution Approach 1:
The patent divides the iris region into multiple sub-regions (e.g., upper iris, lower iris, left iris, right iris) and extracts features from each sub-region separately. This segmentation allows the system to use a fixed set of sub-regions for ensemble learning, eliminating the need for trial and error in selecting regions, thereby reducing learning costs while maintaining recognition performance
Solution Approach 2:
The patent pre-defines a set of sub-regions and their corresponding feature extraction methods before actual authentication occurs. By preparing the feature extraction framework in advance with predetermined sub-regions, the system avoids time-consuming trial and error during the learning phase, reducing learning costs while ensuring reliable recognition
2Measurement precision
If high-resolution images are used for iris recognition, then recognition accuracy improves, but equipment costs and image capture conditions become stricter
Solution Approach 1:
The patent segments the iris region into multiple sub-regions and extracts features from each, then combines these features for authentication. This approach allows the system to achieve high recognition accuracy using standard-resolution images by processing multiple sub-regions, rather than requiring expensive high-resolution imaging equipment
Solution Approach 2:
The patent merges feature vectors from multiple sub-regions (upper, lower, left, right iris regions) to create a comprehensive feature representation for authentication. By combining information from multiple sub-regions processed through standard imaging equipment, the system achieves recognition accuracy previously requiring expensive high-resolution equipment
Data Source
AI summary
Multiple different sub-regions are selected from among sub-regions including at least a portion of an iris region based on features of an eye of a target included in an acquired image. Feature quantities of the respective different sub-regions are calculated. Similarity is calculated between pre-stored features and the respective different sub-regions based on the features of the respective different sub-regions and the pre-stored features. The target is authenticated based on the similarity of the respective different sub-regions.