Sensor-Specific Fingerprint Enhancement for Flexible Displays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Obtaining satisfactory fingerprint image data from a fingerprint sensor deployed in a flexible display device is challenging due to differences in display stack layers, leading to lower quality image capture by some sensors and increased false rejection rates.
Innovation Solution
A control system processes fingerprint image data differently based on the sensor used, enhancing lower quality data using a trained image enhancement machine learning model, such as a generative adversarial network, to transform it to resemble higher quality data for accurate authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fingerprint sensor is deployed in a flexible display device with different display stack layers, then the device can maintain flexibility and display functionality, but the fingerprint image data quality deteriorates due to variations in sensor performance across different panels
Solution Approach 1:
The system applies different processing strategies to fingerprint images based on their source panel. Images from the first panel are processed with standard authentication, while images from the second panel undergo enhanced processing including machine learning-based quality improvement. This local differentiation resolves the contradiction by adapting the quality handling to the specific sensor's characteristics.
Solution Approach 2:
The system changes the processing parameters applied to fingerprint images based on the sensor origin. For lower quality images from the second panel, the system adjusts parameters such as applying quality enhancement algorithms, modifying authentication thresholds, or using different matching criteria to compensate for the reduced image quality while maintaining authentication accuracy.
2Productivity
If fingerprint image data from lower quality sensors is used directly for authentication, then the processing speed is maintained, but the false rejection rate increases
Solution Approach 1:
The system performs preliminary quality assessment of fingerprint images before authentication. Images from the second panel are pre-processed with quality enhancement techniques, and the system determines in advance which images require enhanced processing. This preliminary action prevents false rejections while maintaining efficient processing by only applying enhanced processing where necessary.
Solution Approach 2:
The system introduces an intermediary quality enhancement process between image capture and authentication. For images from the second panel, a machine learning-based enhancement model acts as an intermediary step that improves image quality before the authentication algorithm processes it, thereby reducing false rejections without significantly impacting overall processing speed.
3Adaptability or versatility
If multiple fingerprint sensors are deployed in different panels, then the device maintains functionality across folded states, but the system complexity increases due to different image qualities from each sensor
Solution Approach 1:
The system segments the fingerprint authentication process into distinct pathways based on the sensor origin. Each panel's sensor has its own processing pipeline - the first panel uses standard processing while the second panel uses enhanced processing. This segmentation manages complexity by creating clear, separate handling rules for each sensor type rather than attempting a unified approach.
Solution Approach 2:
The system implements a universal authentication framework that can handle multiple sensor types through a common interface. The control system universally determines the source panel, selects the appropriate processing pathway, and manages both sensor types through unified code that handles quality determination and processing selection, thereby managing complexity through multi-functional design.
Data Source
AI summary
Some methods may involve receiving fingerprint image data. Some methods may involve determining whether the fingerprint image data was captured by a first fingerprint sensor of a device or a second fingerprint sensor of the device. Some methods may involve obtaining an enhanced version of the fingerprint image data based on a determination that the fingerprint image data was captured by the second fingerprint sensor of the device. Some methods may involve providing the enhanced version of the fingerprint image data for authentication. In some examples, obtaining the enhanced version of the fingerprint image data may involve an image enhancement machine learning model. In some examples, prior to obtaining the enhanced version of the fingerprint image data, a determination is made that the fingerprint image data captured by the second fingerprint sensor fails to satisfy a quality threshold.