SlideShare a Scribd company logo
1 of 18
ICDP 2011
Latent Fingerprint Segmentation using
Ridge Template Correlation
Nathan Short, A. Lynn Abbott, Michael S. Hsiao,
Edward A. Fox
Virginia Tech
October 11th, 2011
Motivation
 Large sample of good
quality features
 Supervised acquisition of
sample fingerprint
 Few good quality
features for matching
 Low quality
 Low fingerprint surface
area
Rolled/Plain Fingerprints Latent Fingerprints
1/29/2015
1/29/2015
*Images from NIST SD27
Latent vs. Plain/Rolled Minutia
Count*
Latent vs. Plain/Rolled Minutia
Count
1/29/2015
Motivation (cont.)
 Automated Fingerprint Identification Systems
(AFIS)
 Minutia based
 Aimed towards Plain/Rolled fingerprint matching
 Large sample size
 Latent fingerprints continue to be encoded
manually
1/29/2015
Motivation (cont.)
 Latent matching
 Recent work has included additional features in
matching process [Jain and Feng]
 minutiae, core points, ridge flow, local quality, ridge
wavelength, and others
 matching results much improved over minutia-only based
methods
 All features are extracted manually from latent prints
for matching
 Quality is subjective
1/29/2015
Fingerprint Identification
 Segment Fingerprint
Image
 Enhance Fingerprint
Ridges
 Find Binary Image
 Find Ridge Skeleton
 Extract Minutiae
 Match Sample
template with
database
1/29/2015
Traditional Segmentation
 Normalize Image
 Min-max
 Remove areas with low variance
 Compute Gradient Image
 Approximate first derivative of normalized image by convolving
with Sobel filter
 Threshold based on average magnitude of gradient
within local blocks
1/29/2015
𝛻 𝐼 =
1
𝑛2
𝑖,𝑗 ∈ 𝐵
𝐺 𝑥 𝑖,𝑗
2
+ 𝐺 𝑦 𝑖,𝑗
2
𝐼 𝑀 = 1 𝑖𝑓 𝛻 𝐼 ≥ 𝑡
0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝐺 𝑥 = 𝑆 𝑋 ∗ 𝐼
𝐺 𝑦 = 𝑆 𝑦 ∗ 𝐼
𝐼 =
I − min(𝐼)
max 𝐼 − min(𝐼)
Traditional Segmentation
 Problems
 Assumes background only contains random noise
 Foreground – structure
 Background – no structure
 Latent prints typically have structured
backgrounds
 Resulting in many spurious minutiae when applying
traditional AFIS feature extraction methods
 Also have similar structured background noise in the
fingerprint region itself
1/29/2015
Segmentation Method
1/29/2015
Input Fingerprint
Image
Input Fingerprint
Image
Normalize Image
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Generate ideal ridge
template
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Generate ideal ridge
template
Adjust template to
image mean and
variance
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Generate ideal ridge
template
Adjust template to
image mean and
variance
Take cross sectional
slice orthogonal to
ridge flow at anchor
point within
foreground region
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Cross-correlation of
cross sectional
region with ideal
template
Generate ideal ridge
template
Adjust template to
image mean and
variance
Take cross sectional
slice orthogonal to
ridge flow at anchor
point within
foreground region
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Cross-correlation of
cross sectional
region with ideal
template
Generate ideal ridge
template
Adjust template to
image mean and
variance
Threshold goodness
of fit score to
determine
foreground region
(quality levels) and
background region
Take cross sectional
slice orthogonal to
ridge flow at anchor
point within
foreground region
Find local ridge
frequency map
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Cross-correlation of
cross sectional
region with ideal
template
Generate ideal ridge
template
Adjust template to
image mean and
variance
Threshold goodness
of fit score to
determine
foreground region
(quality levels) and
background region
Segmented
Fingerprint Image
Take cross sectional
slice orthogonal to
ridge flow at anchor
point within
foreground region
Find local ridge
frequency map
Repeatforallblocksinfingerprintregion
Generate Ideal
Template
Input Fingerprint
Image
Normalize Image
Threshold
normalized
intensities to find
initial foreground
region
Cross-correlation of
cross sectional
region with ideal
template
Generate ideal ridge
template
Adjust template to
image mean and
variance
Threshold goodness
of fit score to
determine
foreground region
(quality levels) and
background region
Segmented
Fingerprint Image
Take cross sectional
slice orthogonal to
ridge flow at anchor
point within
foreground region
Find local ridge
frequency map
Ridge Template Generation
 “Ideal” Ridge Template
 Modeled by
𝑇𝑖 = sin 2𝜋𝑓𝑑𝑖 −
𝜋
2
= −cos 2𝜋𝑓𝑑𝑖, , ∀𝑗
 Adjust normalized template to mean and variance of image by
𝑇𝑖 = 𝜎𝑖𝑚𝑔 ∙ 𝑇𝑖 + 𝜇𝑖𝑚𝑔, ∀𝑗
𝑓 𝑑13 = 3
Observed 𝑥-signature Ideal 𝑥-signature
1/29/2015
Segmentation Results
1/29/2015
Segmentation Results (cont.)
Fingerprint Area
(% of total
Image)
False Negatives (% of true
minutiae labelled as
background)
NBIS 60.7 1.41
P1 60.7 0.29
P2 33.6 1.47
P3 45.2 0.69
1/29/2015
Line Detection
 Latent fingerprint matching (Jain and Feng)
 Ridge flow direction
 Negative cost associated with ridge directions that do not match
 lines which dominate the local ridge flow direction, decrease
match score
 Detect lines and remove from directional flow computation
1/29/2015
Line Detection
 Hough-based approach
 A line passing through a point (𝑥, 𝑦), 𝑦 = 𝑚𝑥 + 𝑏 is represented in
Hough space as
𝑟 = 𝑥𝑐𝑜𝑠(𝜃) + 𝑦𝑠𝑖𝑛(𝜃)
 Collinear spatial points are represented by intersecting curves in
Hough space
 Accumulator is used to find highest frequency parameters, (𝑟, 𝜃),
corresponding to points occurring in image
1/29/2015
Line Detection Results
1/29/2015
Future Work
1/29/2015
 Use classifier to determine
background/foreground and quality, instead of
threshold
 Adjust template for ridge thickness
 Performance results with refined directional
map
 Detect and remove errors caused by text in
background
Thank you!
 Questions?

More Related Content

Similar to ICDP 2011

Pre emphasis on data for an adaptive fingerprint image enhancement
Pre emphasis on data for an adaptive fingerprint image enhancementPre emphasis on data for an adaptive fingerprint image enhancement
Pre emphasis on data for an adaptive fingerprint image enhancementIAEME Publication
 
50409621003 fingerprint recognition system-ppt
50409621003  fingerprint recognition system-ppt50409621003  fingerprint recognition system-ppt
50409621003 fingerprint recognition system-pptMohankumar Ramachandran
 
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)Jordi Arnabat
 
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...theijes
 
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...IOSR Journals
 
Comparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesComparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesMade Artha
 
Comparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesComparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesMade Artha
 
Hybrid fingerprint matching algorithm for high accuracy and reliability
Hybrid fingerprint matching algorithm for high accuracy and reliabilityHybrid fingerprint matching algorithm for high accuracy and reliability
Hybrid fingerprint matching algorithm for high accuracy and reliabilityeSAT Publishing House
 
A Review on Edge Detection Algorithms in Digital Image Processing Applications
A Review on Edge Detection Algorithms in Digital Image Processing ApplicationsA Review on Edge Detection Algorithms in Digital Image Processing Applications
A Review on Edge Detection Algorithms in Digital Image Processing Applicationsrahulmonikasharma
 
Enhanced Latent Fingerprint Segmentation through Dictionary Based Approach
Enhanced Latent Fingerprint Segmentation through Dictionary Based ApproachEnhanced Latent Fingerprint Segmentation through Dictionary Based Approach
Enhanced Latent Fingerprint Segmentation through Dictionary Based ApproachEditor IJMTER
 
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...IJERA Editor
 
Positioning And Orientation In Indoor Environments Using Camera 2003
Positioning And Orientation In Indoor Environments Using Camera 2003Positioning And Orientation In Indoor Environments Using Camera 2003
Positioning And Orientation In Indoor Environments Using Camera 2003Pradeep Thomas Thundiyil
 
Review Paper on Image Denoising Techniques
Review Paper  on Image Denoising TechniquesReview Paper  on Image Denoising Techniques
Review Paper on Image Denoising TechniquesIRJET Journal
 
Remotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acmRemotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acmKriti Bajpai
 
A Review Paper on Fingerprint Image Enhancement with Different Methods
A Review Paper on Fingerprint Image Enhancement with Different MethodsA Review Paper on Fingerprint Image Enhancement with Different Methods
A Review Paper on Fingerprint Image Enhancement with Different MethodsIJMER
 
A binarization technique for extraction of devanagari text from camera based ...
A binarization technique for extraction of devanagari text from camera based ...A binarization technique for extraction of devanagari text from camera based ...
A binarization technique for extraction of devanagari text from camera based ...sipij
 
Spatial enhancement techniques
Spatial enhancement techniquesSpatial enhancement techniques
Spatial enhancement techniquesAakanchaAnand
 

Similar to ICDP 2011 (20)

Pre emphasis on data for an adaptive fingerprint image enhancement
Pre emphasis on data for an adaptive fingerprint image enhancementPre emphasis on data for an adaptive fingerprint image enhancement
Pre emphasis on data for an adaptive fingerprint image enhancement
 
Dp34707712
Dp34707712Dp34707712
Dp34707712
 
50409621003 fingerprint recognition system-ppt
50409621003  fingerprint recognition system-ppt50409621003  fingerprint recognition system-ppt
50409621003 fingerprint recognition system-ppt
 
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)
UserDirectedAnalysis of Scanned Images (DOCENG 03 talk)
 
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...
Smqt Based Fingerprint Enhancement And Encryption For Border Crossing Securit...
 
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...
Developmentof Image Enhancement and the Feature Extraction Techniques on Rura...
 
Comparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesComparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print images
 
Comparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print imagesComparative study of various enhancement techniques for finger print images
Comparative study of various enhancement techniques for finger print images
 
Hybrid fingerprint matching algorithm for high accuracy and reliability
Hybrid fingerprint matching algorithm for high accuracy and reliabilityHybrid fingerprint matching algorithm for high accuracy and reliability
Hybrid fingerprint matching algorithm for high accuracy and reliability
 
A Review on Edge Detection Algorithms in Digital Image Processing Applications
A Review on Edge Detection Algorithms in Digital Image Processing ApplicationsA Review on Edge Detection Algorithms in Digital Image Processing Applications
A Review on Edge Detection Algorithms in Digital Image Processing Applications
 
Enhanced Latent Fingerprint Segmentation through Dictionary Based Approach
Enhanced Latent Fingerprint Segmentation through Dictionary Based ApproachEnhanced Latent Fingerprint Segmentation through Dictionary Based Approach
Enhanced Latent Fingerprint Segmentation through Dictionary Based Approach
 
[IJET-V2I2P6] Authors:Atul Ganbawle , Prof J.A. Shaikh
[IJET-V2I2P6] Authors:Atul Ganbawle , Prof J.A. Shaikh[IJET-V2I2P6] Authors:Atul Ganbawle , Prof J.A. Shaikh
[IJET-V2I2P6] Authors:Atul Ganbawle , Prof J.A. Shaikh
 
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...
Analysis of Image Fusion Techniques for fingerprint Palmprint Multimodal Biom...
 
Positioning And Orientation In Indoor Environments Using Camera 2003
Positioning And Orientation In Indoor Environments Using Camera 2003Positioning And Orientation In Indoor Environments Using Camera 2003
Positioning And Orientation In Indoor Environments Using Camera 2003
 
06_features_slides.pdf
06_features_slides.pdf06_features_slides.pdf
06_features_slides.pdf
 
Review Paper on Image Denoising Techniques
Review Paper  on Image Denoising TechniquesReview Paper  on Image Denoising Techniques
Review Paper on Image Denoising Techniques
 
Remotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acmRemotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acm
 
A Review Paper on Fingerprint Image Enhancement with Different Methods
A Review Paper on Fingerprint Image Enhancement with Different MethodsA Review Paper on Fingerprint Image Enhancement with Different Methods
A Review Paper on Fingerprint Image Enhancement with Different Methods
 
A binarization technique for extraction of devanagari text from camera based ...
A binarization technique for extraction of devanagari text from camera based ...A binarization technique for extraction of devanagari text from camera based ...
A binarization technique for extraction of devanagari text from camera based ...
 
Spatial enhancement techniques
Spatial enhancement techniquesSpatial enhancement techniques
Spatial enhancement techniques
 

Recently uploaded

TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024The Digital Insurer
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businesspanagenda
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyKhushali Kathiriya
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesBoston Institute of Analytics
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoffsammart93
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024The Digital Insurer
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsJoaquim Jorge
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...apidays
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 

Recently uploaded (20)

TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
HTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation StrategiesHTML Injection Attacks: Impact and Mitigation Strategies
HTML Injection Attacks: Impact and Mitigation Strategies
 
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot TakeoffStrategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024Manulife - Insurer Innovation Award 2024
Manulife - Insurer Innovation Award 2024
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 

ICDP 2011

  • 1. ICDP 2011 Latent Fingerprint Segmentation using Ridge Template Correlation Nathan Short, A. Lynn Abbott, Michael S. Hsiao, Edward A. Fox Virginia Tech October 11th, 2011
  • 2. Motivation  Large sample of good quality features  Supervised acquisition of sample fingerprint  Few good quality features for matching  Low quality  Low fingerprint surface area Rolled/Plain Fingerprints Latent Fingerprints 1/29/2015
  • 3. 1/29/2015 *Images from NIST SD27 Latent vs. Plain/Rolled Minutia Count*
  • 4. Latent vs. Plain/Rolled Minutia Count 1/29/2015
  • 5. Motivation (cont.)  Automated Fingerprint Identification Systems (AFIS)  Minutia based  Aimed towards Plain/Rolled fingerprint matching  Large sample size  Latent fingerprints continue to be encoded manually 1/29/2015
  • 6. Motivation (cont.)  Latent matching  Recent work has included additional features in matching process [Jain and Feng]  minutiae, core points, ridge flow, local quality, ridge wavelength, and others  matching results much improved over minutia-only based methods  All features are extracted manually from latent prints for matching  Quality is subjective 1/29/2015
  • 7. Fingerprint Identification  Segment Fingerprint Image  Enhance Fingerprint Ridges  Find Binary Image  Find Ridge Skeleton  Extract Minutiae  Match Sample template with database 1/29/2015
  • 8. Traditional Segmentation  Normalize Image  Min-max  Remove areas with low variance  Compute Gradient Image  Approximate first derivative of normalized image by convolving with Sobel filter  Threshold based on average magnitude of gradient within local blocks 1/29/2015 𝛻 𝐼 = 1 𝑛2 𝑖,𝑗 ∈ 𝐵 𝐺 𝑥 𝑖,𝑗 2 + 𝐺 𝑦 𝑖,𝑗 2 𝐼 𝑀 = 1 𝑖𝑓 𝛻 𝐼 ≥ 𝑡 0 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒 𝐺 𝑥 = 𝑆 𝑋 ∗ 𝐼 𝐺 𝑦 = 𝑆 𝑦 ∗ 𝐼 𝐼 = I − min(𝐼) max 𝐼 − min(𝐼)
  • 9. Traditional Segmentation  Problems  Assumes background only contains random noise  Foreground – structure  Background – no structure  Latent prints typically have structured backgrounds  Resulting in many spurious minutiae when applying traditional AFIS feature extraction methods  Also have similar structured background noise in the fingerprint region itself 1/29/2015
  • 10. Segmentation Method 1/29/2015 Input Fingerprint Image Input Fingerprint Image Normalize Image Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Generate ideal ridge template Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Generate ideal ridge template Adjust template to image mean and variance Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Generate ideal ridge template Adjust template to image mean and variance Take cross sectional slice orthogonal to ridge flow at anchor point within foreground region Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Cross-correlation of cross sectional region with ideal template Generate ideal ridge template Adjust template to image mean and variance Take cross sectional slice orthogonal to ridge flow at anchor point within foreground region Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Cross-correlation of cross sectional region with ideal template Generate ideal ridge template Adjust template to image mean and variance Threshold goodness of fit score to determine foreground region (quality levels) and background region Take cross sectional slice orthogonal to ridge flow at anchor point within foreground region Find local ridge frequency map Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Cross-correlation of cross sectional region with ideal template Generate ideal ridge template Adjust template to image mean and variance Threshold goodness of fit score to determine foreground region (quality levels) and background region Segmented Fingerprint Image Take cross sectional slice orthogonal to ridge flow at anchor point within foreground region Find local ridge frequency map Repeatforallblocksinfingerprintregion Generate Ideal Template Input Fingerprint Image Normalize Image Threshold normalized intensities to find initial foreground region Cross-correlation of cross sectional region with ideal template Generate ideal ridge template Adjust template to image mean and variance Threshold goodness of fit score to determine foreground region (quality levels) and background region Segmented Fingerprint Image Take cross sectional slice orthogonal to ridge flow at anchor point within foreground region Find local ridge frequency map
  • 11. Ridge Template Generation  “Ideal” Ridge Template  Modeled by 𝑇𝑖 = sin 2𝜋𝑓𝑑𝑖 − 𝜋 2 = −cos 2𝜋𝑓𝑑𝑖, , ∀𝑗  Adjust normalized template to mean and variance of image by 𝑇𝑖 = 𝜎𝑖𝑚𝑔 ∙ 𝑇𝑖 + 𝜇𝑖𝑚𝑔, ∀𝑗 𝑓 𝑑13 = 3 Observed 𝑥-signature Ideal 𝑥-signature 1/29/2015
  • 13. Segmentation Results (cont.) Fingerprint Area (% of total Image) False Negatives (% of true minutiae labelled as background) NBIS 60.7 1.41 P1 60.7 0.29 P2 33.6 1.47 P3 45.2 0.69 1/29/2015
  • 14. Line Detection  Latent fingerprint matching (Jain and Feng)  Ridge flow direction  Negative cost associated with ridge directions that do not match  lines which dominate the local ridge flow direction, decrease match score  Detect lines and remove from directional flow computation 1/29/2015
  • 15. Line Detection  Hough-based approach  A line passing through a point (𝑥, 𝑦), 𝑦 = 𝑚𝑥 + 𝑏 is represented in Hough space as 𝑟 = 𝑥𝑐𝑜𝑠(𝜃) + 𝑦𝑠𝑖𝑛(𝜃)  Collinear spatial points are represented by intersecting curves in Hough space  Accumulator is used to find highest frequency parameters, (𝑟, 𝜃), corresponding to points occurring in image 1/29/2015
  • 17. Future Work 1/29/2015  Use classifier to determine background/foreground and quality, instead of threshold  Adjust template for ridge thickness  Performance results with refined directional map  Detect and remove errors caused by text in background

Editor's Notes

  1. Accidental friction ridge skin impression left on a surface (crime scene) Typically not visible, made visible by chemicals like powders ninhydrin then photographed or lifted with adhesive
  2. Latent 20.5 (16) Plain 106.3 (80)
  3. 99.4% rank-one identification rate (10,000 images) 54% rank-one identification rate (40 mil) “Lights out” “Semi-Lights Out”
  4. Traditionally been used for classification, but not used in matching
  5. 99.4% rank-one identification rate (10,000 images) 54% rank-one identification rate (40 mil)
  6. Intensity range between [0 1]
  7. - D_i is distance from closest ridge center, 0<=d_i<=f/2 - T_i (-1, 1)
  8. Base – 0.307 vs 0.6924 Prop – 0.319 vs 0.68 Trad – 0.338 vs 0.662