Fingerprint Image Stitching via Edge Detection and Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fingerprint recognition systems face challenges in reconstructing complete fingerprint images due to sensor limitations and image artifacts like blurring and motion, which require substantial processing power and time, potentially frustrating users.
Innovation Solution
The implementation of techniques that examine and correct portions of fingerprint images for blurring and motion artifacts, optimizing image stitching to minimize edge discontinuities and errors, while balancing image quality and processing time through target signal-to-noise ratios and minimal acceptable signal-to-noise ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If substantial processing power is applied to reconstruct complete fingerprint images from portions and correct artifacts, then image quality is improved, but processing time increases
Solution Approach 1:
The fingerprint image is divided into multiple portions or swatches that are captured separately by the sensor. Each portion is processed independently for artifact correction, allowing parallel processing and reducing overall processing time while maintaining image quality through targeted correction of each segment.
Solution Approach 2:
Instead of applying full processing to entire fingerprint images, the system applies correction only to specific portions that contain artifacts or require improvement. This partial action approach reduces computational load and processing time while maintaining necessary image quality through selective correction of problem areas.
2Area of stationary object
If fingerprint sensor size is reduced for device integration, then device compactness is improved, but complete fingerprint capture capability deteriorates
Solution Approach 1:
The small sensor captures multiple smaller portions of the fingerprint rather than attempting to capture the entire fingerprint in one large sensor. These multiple portions are then stitched together computationally to reconstruct the complete fingerprint image, enabling complete capture capability with a compact sensor.
Solution Approach 2:
The system transitions from a single-dimension approach (one large sensor capturing the entire fingerprint) to a multi-dimensional approach (multiple small sensors or multiple passes capturing portions that are then combined). This dimensional change allows complete fingerprint capture while maintaining sensor compactness.
3Area of stationary object
If multiple portions of fingerprint images are captured and stitched together, then complete fingerprint coverage is improved, but edge discontinuity errors increase
Solution Approach 1:
The system performs preliminary processing on each fingerprint portion before stitching, including artifact correction and edge preparation. By preparing each portion in advance with proper alignment and correction, the subsequent stitching process produces smoother transitions and reduces edge discontinuity errors in the final composite image.
Solution Approach 2:
The stitching process incorporates feedback mechanisms where edge characteristics from adjacent portions are analyzed and used to adjust the alignment and blending of neighboring segments. This feedback-driven approach continuously optimizes the stitching to minimize visible discontinuities and maintain edge continuity across the complete fingerprint image.
Data Source
AI summary
An error correction method and a mobile communication device incorporating the error correction method. The error correction method begins with receiving fingerprint image information that includes a set of swatches. A first one of the received swatches is matched to a second one of the swatches at an edge. A measure of fingerprint image manipulation is determined responsive to an amount of processing to perform and an amount of fingerprint image error. Then the fingerprint image information is manipulated in response to an optimum value of the measure.