Fingerprint Image Filtering with Median and Non-Local Means Denoising
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fingerprint images often suffer from noise and imperfections due to imaging limitations and skin characteristics, making it difficult to extract reliable features for identification purposes.
Innovation Solution
A method involving preliminary filtering using median filtering and non-local means denoising, followed by normalization, brightness adjustment, and segmentation processes to produce a refined fingerprint image, which includes applying local mean and variance filters to create a background mask and performing post-processing steps like dilation, noise removal, and filling boundary holes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If median filtering and non-local means denoising are applied to the input fingerprint image, then noise is reduced and image quality is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies preliminary filtering using median filtering and non-local means denoising algorithms before subsequent processing steps. This preliminary action removes noise early in the pipeline, preventing it from propagating through later stages and reducing the computational burden of subsequent operations.
Solution Approach 2:
The patent divides the filtering process into distinct stages: preliminary filtering (median filtering), denoising (non-local means), normalization, and final filtering. This segmentation allows each stage to focus on specific tasks, optimizing both performance and efficiency.
2Reliability
If multiple filtering operations (median filtering, non-local means, local mean filter, local variance filter) are applied sequentially, then noise is effectively removed and background segmentation is improved, but the number of processing steps and computational load increase
Solution Approach 1:
The patent segments the filtering process into distinct functional modules: preliminary filtering, denoising, normalization, background segmentation using local mean and variance filters, and post-processing. Each module performs a specific function, making the complex process manageable and optimizable.
Solution Approach 2:
The patent combines multiple filtering operations (median filtering, non-local means, local mean filter, local variance filter) into a unified processing pipeline where each filter builds upon the previous one, creating a comprehensive noise removal and segmentation system.
3Measurement precision
If brightness adjustment and normalization are performed on the fingerprint image, then image intensity values are optimized for feature extraction, but processing complexity and computational time increase
Solution Approach 1:
The patent applies brightness adjustment and normalization operations that transform image intensity values to optimize the dynamic range and contrast. These parameter changes enhance the visibility of fingerprint features, making them more suitable for subsequent analysis and identification.
Data Source
AI summary
A method of filtering a fingerprint image is provided. The method includes: receiving an input fingerprint image; preliminarily filtering the input fingerprint image to obtain a preprocessed image; normalizing and then filtering the preprocessed image to obtain a denoised image; filtering and then segmenting the preprocessed image to obtain a background mask; postprocessing the background mask to obtain a final mask; and applying the final mask to the denoised image to obtain an output fingerprint image.