Fingerprint Recognition Image Segmentation for Mixed Dry-Wet States
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fingerprint recognition systems face reduced recognition rates and failures when both dry and wet states exist simultaneously, as they rely on a single unlocking mode.
Innovation Solution
A fingerprint recognition device and method that classifies a fingerprint image into high-quality and low-quality areas based on grayscale value ratios, adaptively sampling more feature points from high-quality areas and fewer from low-quality areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single fingerprint unlocking mode is used, then the device complexity is reduced, but the recognition rate deteriorates when both dry and wet states exist simultaneously
Solution Approach 1:
The fingerprint image is divided into multiple sub-areas, and each sub-area is classified into high-quality or low-quality areas based on grayscale value ratios. This segmentation allows the system to handle different fingerprint states (dry and wet) within the same image by processing each region according to its quality, thereby improving recognition rate without requiring multiple separate unlocking modes.
Solution Approach 2:
Different sampling strategies are applied to different regions of the fingerprint image based on their quality. High-quality areas receive different processing than low-quality areas, allowing the system to optimize recognition for each region's characteristics. This local quality approach enables the system to handle mixed dry-wet states effectively while maintaining a single unified unlocking mode.
2Measurement precision
If feature sampling is performed uniformly across all areas, then the processing is simple, but recognition accuracy deteriorates due to including low-quality features from blurry regions
Solution Approach 1:
The fingerprint image is divided into multiple sub-areas for individual quality assessment. Each sub-area is evaluated based on grayscale value ratios to determine whether it represents a high-quality or low-quality region. This segmentation enables selective feature sampling where only high-quality areas contribute features to the recognition process, improving accuracy without requiring complex uniform processing across the entire image.
Solution Approach 2:
The system applies different processing standards to different regions based on their quality characteristics. High-quality areas with clear fingerprint patterns are identified and sampled for feature extraction, while low-quality areas with blurry or indistinct patterns are excluded or given different processing. This local quality approach ensures that only reliable features contribute to recognition, improving accuracy while keeping processing straightforward through quality-based classification.
Data Source
AI summary
A fingerprint recognition device includes a memory and a processor. The memory is configured to store at least one command. The processor is configured to read the at least one command to execute following steps: dividing a fingerprint image into a plurality of sub-areas; classifying the plurality of sub-areas into a plurality of high-quality areas and a plurality of low-quality areas according to a grayscale value ratio of the plurality of sub-areas; obtaining a first feature sample of the plurality of high-quality areas; obtaining a second feature sample of the plurality of low-quality areas, wherein a number of the first feature sample is larger than a number of the second feature sample; and outputting the first feature sample and the second feature sample.