Fingerprint Template Compensation for Small Sensor Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The small size of fingerprint modules in terminals leads to low recognition rates due to insufficient fingerprint image area in registered templates, requiring multiple inputs and resulting in poor quality for inexperienced users.
Innovation Solution
A method and device that improve the fingerprint template by retrieving feature information, determining matching degrees, and adding compensation images to the template if the matching exceeds a preset threshold, enhancing the recognition process without intersecting with existing template information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If the fingerprint module size is reduced to meet terminal appearance requirements, then the terminal equipment size is decreased, but the fingerprint image area in the registered template becomes overly small resulting in low recognition rate
Solution Approach 1:
The patent transitions from a two-dimensional fingerprint image to a three-dimensional fingerprint model by extracting three-dimensional feature information (surface height, curvature, texture) in addition to traditional two-dimensional image data. This dimensional expansion allows the system to capture comprehensive fingerprint characteristics within the limited sensor area, thereby maintaining high recognition accuracy despite the small module size.
Solution Approach 2:
The patent changes the parameter representation from traditional binary image data to multi-dimensional feature parameters including surface height, curvature radius, and texture characteristics. By transforming the fingerprint data into these enhanced parameters, the system can store more informative data in the limited template space, improving recognition reliability without increasing the physical module size.
2Reliability
If multiple fingerprint inputs are required to increase the fingerprint image area, then the recognition rate improves, but the operation complexity and time consumption increase
Solution Approach 1:
The patent performs preliminary extraction and processing of three-dimensional feature information during the initial fingerprint capture, rather than requiring multiple sequential inputs. By pre-extracting comprehensive feature data (surface, curvature, texture) in a single capture event, the system eliminates the need for repeated registration attempts, thereby reducing registration time while maintaining high recognition accuracy.
Solution Approach 2:
The patent implements continuous feature extraction and template updating during the fingerprint capture process. Instead of discrete registration steps, the system continuously processes fingerprint features and updates the template in real-time, allowing the registration to be completed in a single continuous action rather than requiring multiple separate inputs from the user.
3Quantity of substance
If multiple fingerprint recordings are performed to improve template quality, then the fingerprint image area increases, but the quality of recorded template remains low for inexperienced users
Solution Approach 1:
The patent applies different processing weights and quality thresholds to different local features of the fingerprint (such as focusing on distinctive ridges and valleys while being more tolerant of minor variations in less critical areas). This localized quality assessment allows the system to accept and process fingerprint data from inexperienced users more effectively, as it doesn't require uniform high quality across the entire fingerprint image, thereby improving overall template quality without requiring multiple recordings.
Data Source
AI summary
A method, device, and terminal for improving a fingerprint template are proposed. The method includes: receiving first feature information of a recorded fingerprint image based on preset fingerprint feature types; determining a degree of matching between the first feature information and second feature information based on the second feature information corresponding to the fingerprint feature types in a registered fingerprint template; determining whether the degree of matching is higher than or equal to a preset compensation threshold value; and acquiring a compensation image having no intersection with the fingerprint template from the fingerprint image if the degree of matching is higher than or equal to the compensation threshold value, and adding the compensation image to the fingerprint template.