SlideShare a Scribd company logo
Hai Le-Hong, Hoa Nguyen-Ngoc, Thanh
Nguyen-Tri
Vietnam National University of Hanoi
A complete fingerprint matching algorithm on
GPU for a large scale identification system
Outline
1. Introduction
2. Adapting Complete Fingerprint Matching to GPU
3. Experiment
4. Conclusion
1. Introduction
 Fingerprints are most used biometrics
features for identification
 Minutiae-based matching is the most
popular approach
The need for speed
 Although state-of-the-art algorithms are very
accurate, but the need for processing speed for
databases with millions fingerprints are very
demanding
(Minutia Cylinder-Code, a state-of-the-art matching
algorithm, takes 3 milliseconds to perform a matching.
So it takes 3 seconds to identify a fingerprint in a
database of 1000 fingerprints)
Methods to increase the speed
 Graphic Processing Unit (GPU) has been proven
to be a very useful tool to accelerate the
processing speed of computationally intensive
algorithms
GPU
 CUDA-enabled GPU consists
of a set of Streaming
Multiprocessors (SM)
 Each SM schedules and
executes threads in groups of
32 parallel threads called
warps
 A warp executes one common
instruction at a time
 If threads of the same warp
take different paths (due to flow
control instructions), they have
to wait for each other
GPU for Fingerprint Identification
 Not all algorithms can be implemented on GPU
easily
 Recently, having reports for applying GPU for
MCC fingerprint matching like the works of
Gutierrez et al, Capelli et al
 Most approaches make use of GPU for the
filtering process. After that, more accurate
matching algorithms on CPU for remaining
fingerprint candidates are used
2. Our Ideas
 Based on a view using 32 minutiae for each
fingerprint is enough for the matching process
 From statistics of FVC 2002 fingerprint
databases, the average number of minutiae of
each fingerprint is 30
 The average number of matches for a genuine
matching is 6
Details of Our Proposal
 We use one block for matching, each block has
32 threads
 Each thread of the block is used to calculate a
column in the similarity matrix of MCC algorithm
and to find the maximum value in that column
3. Experiment
 We carried out all the experiments on a GTX
GPU, an NVIDIA GeForce GTX 680 with 1536
CUDA cores, Kepler Architecture and 2GB of
memory
 FVC 2002 database was scaled to different
database sizes (ranging from 10000 to 200 000)
Execution time of the first ten queries
with different database sizes
DB size Time (ms) Throughput (KMPS)
10000 58 1724
50000 284 1760
100000 567 1763
150000 850 1764
200000 1105 1809
Speed
 At larger DB sizes, throughput of the proposed
algorithm is stable at 1.8 millions matches per
second
 It is higher than the result reported for previously
published GPU algorithm, which gains 55.7
KMPS on a single GPU device, which is the same
as our device
Accuracy
 We achieved an EER of 1.34% against a 1.26%
of MCC-baseline
 These are certainly minor differences and can be
accepted in real world applications
4. Conclusion
 Our approach uses all minutiae for calculating
cylinders, but choosing 32 minutiae for matching
process, that makes the approach actually fit well
with the GPU computing architecture
 The proposed method does not affect the
accuracy of the original algorithm. The speed of
the adapting algorithm gains state-of-the-art
result
 The proposed approach can be easily scaled-up
thus makes it possible to implement large-scale
fingerprint identification systems with inexpensive
hardware

More Related Content

Similar to A complete fingerprint matching algorithm on gpu

The influence of data size on a high-performance computing memetic algorithm ...
The influence of data size on a high-performance computing memetic algorithm ...The influence of data size on a high-performance computing memetic algorithm ...
The influence of data size on a high-performance computing memetic algorithm ...
journalBEEI
 
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
EUDAT
 
Area Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips ProcessorArea Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips Processor
IOSR Journals
 
Lecture1
Lecture1Lecture1
Lecture1
Asad Abbas
 
Structure And Role Of The Processor
Structure And Role Of The ProcessorStructure And Role Of The Processor
Structure And Role Of The Processor
py7rjs
 
Comprehensive Performance Evaluation on Multiplication of Matrices using MPI
Comprehensive Performance Evaluation on Multiplication of Matrices using MPIComprehensive Performance Evaluation on Multiplication of Matrices using MPI
Comprehensive Performance Evaluation on Multiplication of Matrices using MPI
ijtsrd
 
Parallel implementation of pulse compression method on a multi-core digital ...
Parallel implementation of pulse compression method on  a multi-core digital ...Parallel implementation of pulse compression method on  a multi-core digital ...
Parallel implementation of pulse compression method on a multi-core digital ...
IJECEIAES
 
Microproccessor technology
Microproccessor technologyMicroproccessor technology
Microproccessor technology
ETC
 
cuTau Leaping
cuTau LeapingcuTau Leaping
cuTau Leaping
Amritesh Srivastava
 
A Parallel Computing-a Paradigm to achieve High Performance
A Parallel Computing-a Paradigm to achieve High PerformanceA Parallel Computing-a Paradigm to achieve High Performance
A Parallel Computing-a Paradigm to achieve High Performance
AM Publications
 
Exploring Gpgpu Workloads
Exploring Gpgpu WorkloadsExploring Gpgpu Workloads
Exploring Gpgpu Workloads
Unai Lopez-Novoa
 
FrackingPaper
FrackingPaperFrackingPaper
FrackingPaper
Collin Purcell
 
Convolutional neural networks for speech controlled prosthetic hands
Convolutional neural networks for speech controlled prosthetic handsConvolutional neural networks for speech controlled prosthetic hands
Convolutional neural networks for speech controlled prosthetic hands
Mohsen Jafarzadeh
 
Intelligent Monitoring
Intelligent MonitoringIntelligent Monitoring
Intelligent Monitoring
Intelie
 
A High Performance Fingerprint Matching System for Large Databases Based on GPU
A High Performance Fingerprint Matching System for Large Databases Based on GPUA High Performance Fingerprint Matching System for Large Databases Based on GPU
A High Performance Fingerprint Matching System for Large Databases Based on GPU
Alpesh Kurhade
 
Chap2 slides
Chap2 slidesChap2 slides
Chap2 slides
ashishmulchandani
 
Efficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda cEfficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda c
csandit
 
Efficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda cEfficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda c
csandit
 
BIL406-Chapter-2-Classifications of Parallel Systems.ppt
BIL406-Chapter-2-Classifications of Parallel Systems.pptBIL406-Chapter-2-Classifications of Parallel Systems.ppt
BIL406-Chapter-2-Classifications of Parallel Systems.ppt
Kadri20
 
Monte Carlo on GPUs
Monte Carlo on GPUsMonte Carlo on GPUs
Monte Carlo on GPUs
fcassier
 

Similar to A complete fingerprint matching algorithm on gpu (20)

The influence of data size on a high-performance computing memetic algorithm ...
The influence of data size on a high-performance computing memetic algorithm ...The influence of data size on a high-performance computing memetic algorithm ...
The influence of data size on a high-performance computing memetic algorithm ...
 
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
High Performance & High Throughput Computing - EUDAT Summer School (Giuseppe ...
 
Area Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips ProcessorArea Optimized Implementation For Mips Processor
Area Optimized Implementation For Mips Processor
 
Lecture1
Lecture1Lecture1
Lecture1
 
Structure And Role Of The Processor
Structure And Role Of The ProcessorStructure And Role Of The Processor
Structure And Role Of The Processor
 
Comprehensive Performance Evaluation on Multiplication of Matrices using MPI
Comprehensive Performance Evaluation on Multiplication of Matrices using MPIComprehensive Performance Evaluation on Multiplication of Matrices using MPI
Comprehensive Performance Evaluation on Multiplication of Matrices using MPI
 
Parallel implementation of pulse compression method on a multi-core digital ...
Parallel implementation of pulse compression method on  a multi-core digital ...Parallel implementation of pulse compression method on  a multi-core digital ...
Parallel implementation of pulse compression method on a multi-core digital ...
 
Microproccessor technology
Microproccessor technologyMicroproccessor technology
Microproccessor technology
 
cuTau Leaping
cuTau LeapingcuTau Leaping
cuTau Leaping
 
A Parallel Computing-a Paradigm to achieve High Performance
A Parallel Computing-a Paradigm to achieve High PerformanceA Parallel Computing-a Paradigm to achieve High Performance
A Parallel Computing-a Paradigm to achieve High Performance
 
Exploring Gpgpu Workloads
Exploring Gpgpu WorkloadsExploring Gpgpu Workloads
Exploring Gpgpu Workloads
 
FrackingPaper
FrackingPaperFrackingPaper
FrackingPaper
 
Convolutional neural networks for speech controlled prosthetic hands
Convolutional neural networks for speech controlled prosthetic handsConvolutional neural networks for speech controlled prosthetic hands
Convolutional neural networks for speech controlled prosthetic hands
 
Intelligent Monitoring
Intelligent MonitoringIntelligent Monitoring
Intelligent Monitoring
 
A High Performance Fingerprint Matching System for Large Databases Based on GPU
A High Performance Fingerprint Matching System for Large Databases Based on GPUA High Performance Fingerprint Matching System for Large Databases Based on GPU
A High Performance Fingerprint Matching System for Large Databases Based on GPU
 
Chap2 slides
Chap2 slidesChap2 slides
Chap2 slides
 
Efficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda cEfficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda c
 
Efficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda cEfficient algorithm for rsa text encryption using cuda c
Efficient algorithm for rsa text encryption using cuda c
 
BIL406-Chapter-2-Classifications of Parallel Systems.ppt
BIL406-Chapter-2-Classifications of Parallel Systems.pptBIL406-Chapter-2-Classifications of Parallel Systems.ppt
BIL406-Chapter-2-Classifications of Parallel Systems.ppt
 
Monte Carlo on GPUs
Monte Carlo on GPUsMonte Carlo on GPUs
Monte Carlo on GPUs
 

Recently uploaded

GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
Neo4j
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
Octavian Nadolu
 
Presentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of GermanyPresentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of Germany
innovationoecd
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
mikeeftimakis1
 
“I’m still / I’m still / Chaining from the Block”
“I’m still / I’m still / Chaining from the Block”“I’m still / I’m still / Chaining from the Block”
“I’m still / I’m still / Chaining from the Block”
Claudio Di Ciccio
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
Zilliz
 
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
Neo4j
 
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
SOFTTECHHUB
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Albert Hoitingh
 
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfUnlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Malak Abu Hammad
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
Safe Software
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
Uni Systems S.M.S.A.
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
Daiki Mogmet Ito
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
Data structures and Algorithms in Python.pdf
Data structures and Algorithms in Python.pdfData structures and Algorithms in Python.pdf
Data structures and Algorithms in Python.pdf
TIPNGVN2
 
Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
Adtran
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
sonjaschweigert1
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
KatiaHIMEUR1
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
SOFTTECHHUB
 
20 Comprehensive Checklist of Designing and Developing a Website
20 Comprehensive Checklist of Designing and Developing a Website20 Comprehensive Checklist of Designing and Developing a Website
20 Comprehensive Checklist of Designing and Developing a Website
Pixlogix Infotech
 

Recently uploaded (20)

GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
GraphSummit Singapore | The Future of Agility: Supercharging Digital Transfor...
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
 
Presentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of GermanyPresentation of the OECD Artificial Intelligence Review of Germany
Presentation of the OECD Artificial Intelligence Review of Germany
 
Introduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - CybersecurityIntroduction to CHERI technology - Cybersecurity
Introduction to CHERI technology - Cybersecurity
 
“I’m still / I’m still / Chaining from the Block”
“I’m still / I’m still / Chaining from the Block”“I’m still / I’m still / Chaining from the Block”
“I’m still / I’m still / Chaining from the Block”
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
 
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
GraphSummit Singapore | Enhancing Changi Airport Group's Passenger Experience...
 
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
Goodbye Windows 11: Make Way for Nitrux Linux 3.5.0!
 
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
Encryption in Microsoft 365 - ExpertsLive Netherlands 2024
 
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfUnlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
 
Essentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FMEEssentials of Automations: The Art of Triggers and Actions in FME
Essentials of Automations: The Art of Triggers and Actions in FME
 
Microsoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdfMicrosoft - Power Platform_G.Aspiotis.pdf
Microsoft - Power Platform_G.Aspiotis.pdf
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
 
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
Data structures and Algorithms in Python.pdf
Data structures and Algorithms in Python.pdfData structures and Algorithms in Python.pdf
Data structures and Algorithms in Python.pdf
 
Pushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 daysPushing the limits of ePRTC: 100ns holdover for 100 days
Pushing the limits of ePRTC: 100ns holdover for 100 days
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
 
Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !Securing your Kubernetes cluster_ a step-by-step guide to success !
Securing your Kubernetes cluster_ a step-by-step guide to success !
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
 
20 Comprehensive Checklist of Designing and Developing a Website
20 Comprehensive Checklist of Designing and Developing a Website20 Comprehensive Checklist of Designing and Developing a Website
20 Comprehensive Checklist of Designing and Developing a Website
 

A complete fingerprint matching algorithm on gpu

  • 1. Hai Le-Hong, Hoa Nguyen-Ngoc, Thanh Nguyen-Tri Vietnam National University of Hanoi A complete fingerprint matching algorithm on GPU for a large scale identification system
  • 2. Outline 1. Introduction 2. Adapting Complete Fingerprint Matching to GPU 3. Experiment 4. Conclusion
  • 3. 1. Introduction  Fingerprints are most used biometrics features for identification  Minutiae-based matching is the most popular approach
  • 4. The need for speed  Although state-of-the-art algorithms are very accurate, but the need for processing speed for databases with millions fingerprints are very demanding (Minutia Cylinder-Code, a state-of-the-art matching algorithm, takes 3 milliseconds to perform a matching. So it takes 3 seconds to identify a fingerprint in a database of 1000 fingerprints)
  • 5. Methods to increase the speed  Graphic Processing Unit (GPU) has been proven to be a very useful tool to accelerate the processing speed of computationally intensive algorithms
  • 6. GPU  CUDA-enabled GPU consists of a set of Streaming Multiprocessors (SM)  Each SM schedules and executes threads in groups of 32 parallel threads called warps  A warp executes one common instruction at a time  If threads of the same warp take different paths (due to flow control instructions), they have to wait for each other
  • 7. GPU for Fingerprint Identification  Not all algorithms can be implemented on GPU easily  Recently, having reports for applying GPU for MCC fingerprint matching like the works of Gutierrez et al, Capelli et al  Most approaches make use of GPU for the filtering process. After that, more accurate matching algorithms on CPU for remaining fingerprint candidates are used
  • 8. 2. Our Ideas  Based on a view using 32 minutiae for each fingerprint is enough for the matching process  From statistics of FVC 2002 fingerprint databases, the average number of minutiae of each fingerprint is 30  The average number of matches for a genuine matching is 6
  • 9. Details of Our Proposal  We use one block for matching, each block has 32 threads  Each thread of the block is used to calculate a column in the similarity matrix of MCC algorithm and to find the maximum value in that column
  • 10. 3. Experiment  We carried out all the experiments on a GTX GPU, an NVIDIA GeForce GTX 680 with 1536 CUDA cores, Kepler Architecture and 2GB of memory  FVC 2002 database was scaled to different database sizes (ranging from 10000 to 200 000)
  • 11. Execution time of the first ten queries with different database sizes DB size Time (ms) Throughput (KMPS) 10000 58 1724 50000 284 1760 100000 567 1763 150000 850 1764 200000 1105 1809
  • 12. Speed  At larger DB sizes, throughput of the proposed algorithm is stable at 1.8 millions matches per second  It is higher than the result reported for previously published GPU algorithm, which gains 55.7 KMPS on a single GPU device, which is the same as our device
  • 13. Accuracy  We achieved an EER of 1.34% against a 1.26% of MCC-baseline  These are certainly minor differences and can be accepted in real world applications
  • 14. 4. Conclusion  Our approach uses all minutiae for calculating cylinders, but choosing 32 minutiae for matching process, that makes the approach actually fit well with the GPU computing architecture  The proposed method does not affect the accuracy of the original algorithm. The speed of the adapting algorithm gains state-of-the-art result  The proposed approach can be easily scaled-up thus makes it possible to implement large-scale fingerprint identification systems with inexpensive hardware

Editor's Notes

  1. Minutiae are the points where the ridge continuity breaks The task in the minutiae-based matching approach is finding the maximum number of matching minutiae pairs in two given fingerprints.
  2. used a different fingerprint database in the experiments, the average number of minutiae of a fingerprint is quite stable.