Iris Authentication Accuracy via Multi-View Image Quality Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing iris authentication technologies suffer from low accuracy due to variations in image quality, such as motion blurring, focus issues, and camera noise, leading to incorrect authentication of users.

Innovation Solution

An information processing apparatus that captures images of the eye from multiple sight line directions, determines image quality, and combines iris authentication results with sight line movement pattern authentication to enhance accuracy, excluding low-quality image data for improved authentication.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If iris authentication is performed based on a captured image, then authentication can be implemented, but authentication accuracy is low due to image quality variations

Engineering Contradiction:
Improveauthentication accuracyVSAvoidiris feature identification accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent segments the authentication process into multiple independent authentication results, each derived from a separate captured image. By dividing the authentication into multiple trials and selecting the result with the highest image quality, the system reduces the impact of poor quality images on overall authentication accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary image quality determination before final authentication result selection. By evaluating image quality metrics in advance, the system can identify and prioritize high-quality images for authentication, ensuring more accurate iris feature extraction and comparison.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If multiple captured images are used for authentication, then authentication accuracy improves, but processing complexity increases

Engineering Contradiction:
Improveauthentication accuracyVSAvoidprocessing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and evaluates specific image quality metrics from each captured image, such as focus quality, noise levels, and iris region clarity. By isolating and measuring these key quality parameters, the system can efficiently compare multiple images and select the best one without processing all image data equally, reducing overall processing complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements a feedback mechanism where image quality determination results from previous processing steps inform the selection of images for authentication. The system uses quality metrics as feedback to prioritize which images to process further, creating an efficient workflow that improves accuracy while managing processing complexity through intelligent decision-making.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3229205B1Information processing device, information processing method, and program
Publication Date: 2021.10.13 SONY GROUP CORP
  • EP3229205B1 patent drawingFigure 1
  • EP3229205B1 patent drawingFigure 2
  • EP3229205B1 patent drawingFigure 3

AI summary

[Object] To provide an information processing apparatus, an information processing method, and a program, which can improve accuracy of authentication based on biometric information of an eye. [Solution] Provided is an information processing apparatus including: a biometric information authentication unit that authenticates biometric information identified from each of a plurality of captured images of an eye of a user of different sight line directions, on the basis of reference images of the eye of the user of the respective sight line directions; and an authentication result combining unit that combines authentication results by the biometric information authentication unit.