Facial Recognition Image Selection for Appearance Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional facial recognition systems in venues like airports face reduced identification accuracy when users change their appearance, such as by wearing hats, glasses, or adjusting their hair, as the comparison between initial and subsequent images leads to decreased matching scores and potential identification failures.
Innovation Solution
A method and device for user identification that acquire and compare images based on a selection metric, selecting the most recent image associated with a previous identification to improve accuracy and reliability, conserve resources, and enhance throughput and latency in authentication processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system compares the initial image with subsequent images acquired after appearance changes, then the identification process can be performed, but the identification accuracy is reduced and identification may become impossible
Solution Approach 1:
The system performs preliminary actions by acquiring multiple images at different stages (initial image, intermediate images, and final images) before the final identification decision is made. These preliminary images are stored and used as reference points for comparison, allowing the system to adapt to appearance changes that occur during the user's journey through the airport.
Solution Approach 2:
The system dynamically selects which images to compare based on the user's appearance changes. Instead of a fixed comparison between initial and final images only, the system can select intermediate images that best match the user's current appearance, making the identification process adaptable to varying conditions.
2Reliability
If the system stores and compares multiple images of the user at different locations, then identification accuracy improves, but device complexity and resource consumption increase
Solution Approach 1:
The identification process is segmented into distinct stages: initial image acquisition, intermediate image acquisition at different locations (baggage drop, security, boarding), and final identification. Each stage uses only the necessary images for that specific purpose, reducing the overall complexity of managing all images simultaneously.
Solution Approach 2:
The system extracts and stores only the essential features and characteristics from each acquired image rather than storing complete high-resolution images. This extraction of key identification features reduces storage requirements and simplifies comparison operations while maintaining identification accuracy.
Data Source
AI summary
A device and method for performing a user identification of a user include acquiring a first image of the user. A second image of the user that is associated with a previous user identification of the user is selected. The first image and the second image are compared, and a user identification is performed based on the comparison of the first image the second image.