Fingerprint Sensor Dynamic-Range Mapping for Trace Discrimination

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Fingerprint sensors are sensitive to skin type, humidity, and lighting conditions, leading to high variability in fingerprint images and difficulty in distinguishing between actual fingerprints and traces left on the sensor surface.

Innovation Solution

A method involving peak and valley value analysis to determine a normalized dynamic range for each pixel, comparing it against thresholds to generate a mask that distinguishes between fingerprints and traces, using a fingerprint sensor and image processing module to refine the image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional fingerprint sensing is used, then fingerprint images can be captured, but the images show high variability due to skin type, humidity, and lighting conditions, making it difficult to distinguish between actual fingerprints and traces

Engineering Contradiction:

Improvediscrimination accuracy between fingerprints and tracesVSAvoidvariability due to skin type, humidity, and lighting conditions

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent transforms the raw image data into a normalized dynamic range map by changing the parameter representation from absolute pixel values to relative dynamic range values. This parameter transformation makes the fingerprint representation invariant to lighting conditions, skin type, and humidity variations, thereby resolving the discrimination accuracy problem

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

Instead of trying to enhance the fingerprint signal directly, the patent inverts the approach by analyzing the dynamic range (difference between peak and valley values) and normalizing it. This inverted approach of focusing on relative differences rather than absolute values eliminates the harmful variability factors

Inventive Principle:

Principle #13The other way round (Inversion)

2Loss of information

If the sensor captures all image data including traces, then complete image information is obtained, but traces are difficult to distinguish from real fingerprints, leading to false alarms

Engineering Contradiction:

Improveinformation completeness of image dataVSAvoidauthentication accuracy

Core Design Contradiction:

Loss of informationVSReliability

Solution Approach 1:

The patent segments the image processing into distinct stages: first capturing complete image data, then calculating peak and valley values for each pixel, computing the normalized dynamic range, and finally applying threshold comparison. This segmentation allows complete information capture while systematically eliminating traces through multi-stage processing

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary representation - the normalized dynamic range map - that serves as a bridge between the raw image data and the final authentication decision. This intermediary transformation preserves useful fingerprint information while filtering out trace information, resolving the reliability issue

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentEP4618048A1Method for processing an image acquired by a fingerprint sensor to discriminate fingers and traces

Publication Date: 2025.09.17 IDEMIA PUBLIC SECURITY FRANCE

Figure 1~2
Figure 3~4
Figure 5

AI summary

It comprises determining (200) a peak value (R) associated with a pixel of the image and a valley value (V) associated with the pixel; calculating (202) a dynamic range (Dn) associated with the pixel as a ratio between a deviation between the peak value (R) and the valley value (V) associated with the pixel, and a reference value being a linear combination of the peak value (R) and the valley value (V) associated with the pixel; comparing (204) the dynamic range associated with the pixel and a threshold (T1); and generating a result associated with the pixel, said result indicating that the pixel shows a finger in view of the fingerprint sensor only if the dynamic range (Dn) is greater than the threshold (T1).