Character Noise Eliminating Apparatus for Latent Fingerprint Images

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods are ineffective in accurately eliminating atypical background noises, such as character noise, from latent fingerprint images, leading to difficulties in enhancing and extracting fingerprint ridgelines, which hinders automated fingerprint matching systems.

Innovation Solution

A character noise eliminating apparatus and method that detects character noise areas, sets density conversion layers inside and outside these areas, and applies local image enhancement using adaptive histogram equalization or adaptive contrast stretch, limiting the reference area to neighboring pixel groups within the same density conversion layer to effectively eliminate character noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If Fourier transformation is employed to eliminate background pattern noise, then periodic background noises can be removed, but character noises with irregular shapes cannot be eliminated effectively and fingerprint ridgeline density deteriorates

Engineering Contradiction:

Improveperiodic background noiseVSAvoidfingerprint ridgeline density

Core Design Contradiction:

Object-affected harmful factorsVSManufacturing precision

Solution Approach 1:

The patent applies local image enhancement (adaptive histogram equalization or adaptive contrast stretch) to specific regions identified as character noise areas. By processing only the problematic regions with appropriate enhancement techniques while preserving other areas, the method removes character noise without degrading the overall fingerprint ridgeline density.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent changes the processing parameters dynamically by detecting character noise areas first, then applying density conversion only to those specific regions. This selective parameter application allows removal of irregular character noises while maintaining fingerprint ridgeline integrity in non-noise areas.

Inventive Principle:

Principle #35Parameter changes

2Object-affected harmful factors

If density conversion is applied to character noise areas, then character noise can be removed, but ridgeline information inside the noise area is eliminated and the image appears artificial

Engineering Contradiction:

Improvecharacter noiseVSAvoidridgeline information

Core Design Contradiction:

Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent extracts and removes only the character noise components from the image while preserving the underlying ridgeline information. By selectively targeting and eliminating the noise elements without affecting the fingerprint structures, the method maintains ridgeline information while removing character noise.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses local image enhancement techniques as an intermediary process between the original image and the final output. This intermediary enhancement step allows noise removal while preserving and even enhancing ridgeline information, avoiding the artificial appearance that results from direct density conversion.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Manufacturing precision

If related art methods are used to enhance fingerprint ridgelines, then ridgeline enhancement can be achieved, but the methods are ineffective when ridgeline directions and periodicities cannot be extracted due to character noise influence

Engineering Contradiction:

Improveridgeline enhancementVSAvoidridgeline extraction accuracy

Core Design Contradiction:

Manufacturing precisionVSReliability

Solution Approach 1:

The patent performs preliminary character noise detection and removal before attempting ridgeline enhancement. By eliminating the character noise interference in advance, the subsequent ridgeline enhancement processes can accurately extract ridgeline directions and periodicities without being distorted by noise, thereby improving both enhancement quality and extraction reliability.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS8014574B2Character noise eliminating apparatus, character noise eliminating method, and character noise eliminating program

Publication Date: 2011.09.06 NEC CORP

AI summary

To provide a noise eliminating apparatus and the like that can eliminate an atypical shaped background noise. A character noise eliminating apparatus includes a character noise area detecting device for detecting a character noise area which is an area corresponding to a character noise from an image, a density conversion area layer determining device for setting a plurality of density conversion area layers inside and outside the character noise area, and a density converting device for setting a neighboring pixel group within the same density conversion area layer as the density conversion area layer to which a target pixel belongs as a reference area of the target pixel, with respect to pixels in the density conversion area layers, and generating a density converted image applying a local image enhancement.