Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/usnistgov/NFIQ2/llms.txt

Use this file to discover all available pages before exploring further.

The LCS (Local Clarity Score) module measures the local clarity of ridge-valley structures in fingerprint images. It analyzes the sharpness and definition of ridge patterns across local regions to assess overall image quality.

​
Overview

Local Clarity Score evaluates how well-defined and clear the ridge-valley structures are in local neighborhoods of the fingerprint image. Unlike frequency-domain methods, LCS operates in the spatial domain to directly measure ridge clarity based on local gradients and contrast patterns.
LCS is particularly sensitive to image sharpness, ridge definition, and local contrast quality. It complements frequency-domain analysis by providing spatial clarity measurements.

​
Class Definition

Header: quality_modules/LCS.h 
class LCS : public Algorithm {
public:
    LCS(const NFIQ2::FingerprintImageData &fingerprintImage);
    virtual ~LCS();
    
    std::string getName() const override;
    static std::vector<std::string> getNativeQualityMeasureIDs();

private:
    std::unordered_map<std::string, double> computeFeatureData(
        const NFIQ2::FingerprintImageData &fingerprintImage);
    
    const int blocksize { Sizes::LocalRegionSquare };
    const double threshold { .1 };
    const int scannerRes { 500 };
    const bool padFlag { false };
};

​
Constructor

​
LCS()

LCS(const NFIQ2::FingerprintImageData &fingerprintImage)
Creates an LCS module instance and computes local clarity features. Parameters:
  • fingerprintImage - Input fingerprint image data at 500 dpi
Throws:
  • NFIQ2::Exception - If the image resolution is not 500 dpi

​
Methods

​
getName()

std::string getName() const override
Returns the module identifier: "LocalClarity" Returns: Module name as string

​
getNativeQualityMeasureIDs()

static std::vector<std::string> getNativeQualityMeasureIDs()
Returns the list of quality measure identifiers produced by this module. Returns: Vector containing feature IDs:
  • LCS_Bin10_0 through LCS_Bin10_9 - 10 histogram bins
  • LCS_Bin10_Mean - Mean local clarity score
  • LCS_Bin10_StdDev - Standard deviation of clarity scores

​
Quality Features

The LCS module generates 12 quality measures based on local clarity analysis:

​
Histogram Bins (0-9)

The local clarity score is computed for each block in the fingerprint image. These scores are then binned into 10 histogram categories using predefined thresholds: 
static double LCSHISTLIMITS[9] = { 0, 0.70, 0.74, 0.77, 0.79, 0.81, 0.83, 0.85, 0.87 };
Each bin represents the proportion of blocks with clarity scores in specific ranges:
  • Lower bins (0-2): Poor clarity, unclear ridge structures
  • Middle bins (3-6): Moderate clarity
  • Higher bins (7-9): Excellent clarity, well-defined ridges

​
Statistical Measures

  • LCS_Bin10_Mean: Average clarity score across all blocks
  • LCS_Bin10_StdDev: Standard deviation indicating clarity consistency
Higher LCS values indicate better ridge clarity and image sharpness. The module requires 500 dpi resolution images.

​
Configuration

The module uses the following default parameters:
ParameterValueDescription
blocksizeSizes::LocalRegionSquareBlock size for local analysis
threshold0.1Minimum threshold for valid regions
scannerRes500Expected scanner resolution (dpi)
padFlagfalsePadding disabled for boundary blocks

​
Usage Example

#include <quality_modules/LCS.h>

NFIQ2::FingerprintImageData fpImage = /* load image */;

try {
    NFIQ2::QualityMeasures::LCS lcs(fpImage);
    
    // Get all feature IDs
    auto featureIDs = NFIQ2::QualityMeasures::LCS::getNativeQualityMeasureIDs();
    
    // Access computed features
    auto features = lcs.getFeatures();
    double meanClarity = features["LCS_Bin10_Mean"];
    double stddev = features["LCS_Bin10_StdDev"];
    
    // Analyze histogram distribution
    for (int i = 0; i < 10; i++) {
        std::string binKey = "LCS_Bin10_" + std::to_string(i);
        double binValue = features[binKey];
        // Process bin values
    }
    
} catch (const NFIQ2::Exception &e) {
    // Handle error
}

​
Algorithm Details

The LCS algorithm:
  1. Divides the fingerprint image into local blocks
  2. For each block:
    • Computes local gradients and contrast measures
    • Analyzes ridge-valley transitions
    • Calculates a clarity score based on pattern sharpness
  3. Creates a histogram distribution of clarity scores
  4. Computes statistical measures (mean and standard deviation)

​
Clarity Score Interpretation

  • 0.0 - 0.70: Very poor clarity, smudged or unclear patterns
  • 0.70 - 0.77: Poor to fair clarity
  • 0.77 - 0.83: Moderate clarity, acceptable for some applications
  • 0.83 - 0.87: Good clarity, clear ridge structures
  • Above 0.87: Excellent clarity, highly defined patterns

​
Relationship to Other Modules

LCS complements other quality modules:
  • FDA: While FDA analyzes frequency content, LCS directly measures spatial clarity
  • OCL: LCS measures clarity while OCL measures orientation certainty
  • Mu: LCS evaluates local sharpness while Mu evaluates overall contrast

Build docs developers (and LLMs) love

Get started for freeTalk to us