Fingerprint Segmentation via Pooled Sub-Images and Region Growing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing fingerprint segmentation algorithms consume excessive computational resources and latency, making them unsuitable for embedded devices with limited resources.

Innovation Solution

A fingerprint segmentation method involving dividing images into sub-images, pooling these sub-images to create a feature map, and using a region growing algorithm with filtering and adaptive thresholds to segment the fingerprint region.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If ridge line orientation computation and complex feature extraction are performed, then segmentation accuracy is improved, but computational complexity increases

Engineering Contradiction:

Improvesegmentation accuracyVSAvoidalgorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The fingerprint image is divided into multiple sub-images (e.g., 8x8 blocks), and each sub-image is processed independently to generate feature values. This segmentation approach reduces the computational complexity of processing the entire image at once while maintaining segmentation accuracy through localized feature extraction.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts only the necessary feature values (minimum gray values) from each sub-image, discarding redundant information. This selective extraction simplifies the feature set used for segmentation, reducing computational complexity while preserving the essential characteristics needed for accurate fingerprint segmentation.

Inventive Principle:

Principle #2Taking out (Extraction)

2Measurement precision

If classifiers such as neural network or SVM are used for foreground-background discrimination, then segmentation accuracy is improved, but computational resource consumption increases

Engineering Contradiction:

Improvediscrimination accuracyVSAvoidcomputational resource consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent replaces expensive classifiers like neural networks and SVM with a simple threshold comparison operation. Instead of using computationally intensive machine learning models, the invention uses a straightforward minimum gray value threshold to discriminate between foreground and background, dramatically reducing computational resource consumption while maintaining acceptable segmentation accuracy.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Measurement precision

If existing fingerprint segmentation algorithms are used, then segmentation accuracy is achieved, but processing time increases

Engineering Contradiction:

Improvesegmentation accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by first dividing the image into sub-images and extracting minimum gray values before the actual segmentation decision. This preprocessing step simplifies the subsequent threshold comparison operation, enabling faster processing while maintaining segmentation accuracy. The preliminary feature extraction reduces the complexity of the final discrimination step.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12525052B2Fingerprint segmentation method and device

Publication Date: 2026.01.13 SAMSUNG ELECTRONICS CO LTD

AI summary

A fingerprint segmentation method and device is provided. The fingerprint segmentation method may include dividing a fingerprint image into a plurality of sub-images, pooling each sub-image to acquire a feature map of the fingerprint image, and segmenting the feature map based on a region growing algorithm to acquire a fingerprint region in the fingerprint image.