Fingerprint Image Region Segmentation for Compact Sensor Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fingerprint recognition technologies face challenges in accurately authenticating users due to poor recognition performance, especially when the sensing area is smaller than the fingerprint, leading to erroneous authentication of authorized and unauthorized users.
Innovation Solution
A method that divides input and registered fingerprint images into multiple regions, compares these regions for phase correlations, generates correction values for rotation and translation transforms, and determines matching probabilities to improve authentication accuracy by enhancing the overlap region's matching score and area analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If the sensing area is reduced to make the device compact, then device size is reduced, but fingerprint recognition accuracy deteriorates
Solution Approach 1:
The patent divides the fingerprint image into multiple regions (first regions and second regions) and performs separate phase correlation calculations for each region. This segmentation allows the system to process limited sensing area data more effectively by focusing on specific sub-regions, thereby maintaining recognition accuracy despite the reduced overall sensing area.
Solution Approach 2:
The patent applies different processing strategies to different regions of the fingerprint image. By performing phase correlation on specific matching regions rather than uniformly processing the entire image, the system optimizes recognition accuracy for the critical overlapping areas while working within the constraints of a compact sensing area.
2Reliability
If traditional full-image comparison is used, then comprehensive matching is achieved, but processing time and computational complexity increase
Solution Approach 1:
The patent segments the fingerprint image into multiple regions and performs phase correlation calculations on selected matching regions rather than comparing the entire image. This segmentation strategy maintains comprehensive matching reliability by focusing on critical sub-regions while significantly reducing the computational burden and processing time.
Solution Approach 2:
The patent extracts and selects specific matching regions from the divided fingerprint image based on phase correlation analysis. By taking out only the most relevant regions for comparison, the system achieves reliable authentication without the need to process the entire image, thereby reducing processing time and computational resources required.
3Measurement precision
If phase correlation is calculated for all regions, then comprehensive matching is achieved, but computational complexity increases
Solution Approach 1:
The patent divides the fingerprint image into multiple regions and performs phase correlation calculations selectively on these divided regions rather than on the entire image at once. This segmentation approach maintains matching precision by analyzing critical sub-regions while reducing the overall computational complexity of the authentication process.
Solution Approach 2:
The patent performs phase correlation calculations on selected matching regions rather than all possible regions. By applying partial action to the most critical sub-regions, the system achieves sufficient matching precision without the excessive computational complexity that would result from analyzing every possible region combination.
Data Source
AI summary
A method for processing fingerprint information includes dividing an input image that corresponds to at least a portion of a user fingerprint into a plurality of first regions; dividing a registered image that had previously been stored into a plurality of second regions; selecting a first matching region from among the plurality of first regions, and selecting a second matching region from among the plurality of second regions, by comparing the plurality of first regions with the plurality of second regions; and matching the registered image with the input image by comparing the first matching region with the second matching region.