Fingerprint Classification Using Regular Expression Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fingerprint classification systems are inefficient and less accurate in classifying digital fingerprint images, requiring more memory and processor-intensive calculations while struggling to achieve high reliability and speed.
Innovation Solution
A fingerprint classification system and method utilizing regular expressions and a LMK classifier model, which enhances digital fingerprint images, extracts dominant orientations, and matches these with pre-defined regular expressions to classify fingerprints into categories like left loop, right loop, whorl loop, and arch loop, using a bank of regular expressions to prioritize matching patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional fingerprint classification systems use traditional classification methods, then they can process fingerprint images, but they require more memory and processor-intensive calculations while operating slower with lower accuracy
Solution Approach 1:
The patent transforms the fingerprint classification approach by changing the parameter representation from raw pixel data to orientation feature vectors derived from ridge flow directions. This parameter transformation enables the use of regular expression matching instead of computationally intensive traditional classification algorithms, significantly reducing processor requirements while improving classification speed and accuracy
Solution Approach 2:
The patent replaces the mechanical computation approach of traditional classification algorithms with a pattern-matching system based on regular expressions. Instead of using heavy mathematical computations and matrix operations, the system uses symbolic pattern matching on orientation sequences, which is computationally lighter and faster while maintaining high accuracy
2Measurement precision
If conventional systems use detailed image processing, then they can extract features, but they consume more memory and computational resources
Solution Approach 1:
The patent extracts only the essential orientation information from fingerprint images by analyzing ridge flow directions and representing them as orientation feature vectors. This extraction approach discards redundant pixel data while retaining the critical structural information needed for classification, thereby achieving high accuracy with minimal memory consumption
Solution Approach 2:
The patent segments the fingerprint image analysis into distinct orientation estimation steps, where the image is divided into blocks and orientation is calculated for each block independently. This segmentation allows efficient processing of large images by breaking them into manageable units, reducing overall memory requirements while maintaining classification precision
Data Source
AI summary
A fingerprint classification system and method for extracting the dominant singularity from a fingerprint image are described. The fingerprint classification system and method receive as an input a digital fingerprint image and the image is preprocessed to generate an enhanced and more accurate image. Feature pattern calculations are performed on the updated image to generate an Orientation Feature Vector. The Orientation Feature Vector is processed using a Regular Expression Machine classifier prediction model to generate a class label for the digital fingerprint image that was input.