Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems
Latent fingerprint identification is one of the most essential identification procedures in criminal investigations. Addressing this task is challenging as (i) it requires analyzing massive databases in reasonable periods and (ii) it is commonly solved by combining different methods with very comple...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9127429/ |
| _version_ | 1829138303584043008 |
|---|---|
| author | Andres J. Sanchez-Fernandez Luis F. Romero Daniel Peralta Miguel Angel Medina-Perez Yvan Saeys Francisco Herrera Siham Tabik |
| author_facet | Andres J. Sanchez-Fernandez Luis F. Romero Daniel Peralta Miguel Angel Medina-Perez Yvan Saeys Francisco Herrera Siham Tabik |
| author_sort | Andres J. Sanchez-Fernandez |
| collection | DOAJ |
| description | Latent fingerprint identification is one of the most essential identification procedures in criminal investigations. Addressing this task is challenging as (i) it requires analyzing massive databases in reasonable periods and (ii) it is commonly solved by combining different methods with very complex data-dependencies, which make fully exploiting heterogeneous CPU-GPU systems very complex. Most efforts in this context focus on improving the accuracy of the approaches and neglect reducing the processing time. Indeed, the most accurate approach was designed for one single thread. This work introduces the fastest methodology for latent fingerprint identification maintaining high accuracy called Asynchronous processing for Latent Fingerprint Identification (ALFI). ALFI fully exploits all the resources of CPU-GPU systems using asynchronous processing and fine-coarse parallelism for analyzing massive databases. Our approach reduces idle times in processing and exploits the inherent parallelism of comparing latent fingerprints to fingerprint impressions. We analyzed the performance of ALFI on Linux and Windows operating systems using the well-known NIST/FVC databases. Experimental results reveal that ALFI is in average 22x faster than the state-of-the-art algorithm, reaching a value of 44.7x for the best-studied case. |
| first_indexed | 2024-12-14T19:14:30Z |
| format | Article |
| id | doaj.art-27a6ba6aff1149c09cd29d1a5d70535a |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T19:14:30Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-27a6ba6aff1149c09cd29d1a5d70535a2022-12-21T22:50:38ZengIEEEIEEE Access2169-35362020-01-01812423612425310.1109/ACCESS.2020.30054769127429Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU SystemsAndres J. Sanchez-Fernandez0https://orcid.org/0000-0001-6743-3570Luis F. Romero1https://orcid.org/0000-0003-2959-2030Daniel Peralta2https://orcid.org/0000-0002-7544-8411Miguel Angel Medina-Perez3https://orcid.org/0000-0003-4511-2252Yvan Saeys4https://orcid.org/0000-0002-0415-1506Francisco Herrera5https://orcid.org/0000-0002-7283-312XSiham Tabik6https://orcid.org/0000-0003-4093-5356Department of Computer Architecture, University of Malaga, Malaga, SpainDepartment of Computer Architecture, University of Malaga, Malaga, SpainDepartment of Applied Mathematics, Computer Science and Statistics, Ghent University, Ghent, BelgiumTecnologico de Monterrey, Atizapán, MéxicoDepartment of Applied Mathematics, Computer Science and Statistics, Ghent University, Ghent, BelgiumAndalusian Research Institute in Data Science and Computational Intelligence, University of Granada, Granada, SpainAndalusian Research Institute in Data Science and Computational Intelligence, University of Granada, Granada, SpainLatent fingerprint identification is one of the most essential identification procedures in criminal investigations. Addressing this task is challenging as (i) it requires analyzing massive databases in reasonable periods and (ii) it is commonly solved by combining different methods with very complex data-dependencies, which make fully exploiting heterogeneous CPU-GPU systems very complex. Most efforts in this context focus on improving the accuracy of the approaches and neglect reducing the processing time. Indeed, the most accurate approach was designed for one single thread. This work introduces the fastest methodology for latent fingerprint identification maintaining high accuracy called Asynchronous processing for Latent Fingerprint Identification (ALFI). ALFI fully exploits all the resources of CPU-GPU systems using asynchronous processing and fine-coarse parallelism for analyzing massive databases. Our approach reduces idle times in processing and exploits the inherent parallelism of comparing latent fingerprints to fingerprint impressions. We analyzed the performance of ALFI on Linux and Windows operating systems using the well-known NIST/FVC databases. Experimental results reveal that ALFI is in average 22x faster than the state-of-the-art algorithm, reaching a value of 44.7x for the best-studied case.https://ieeexplore.ieee.org/document/9127429/BiometricsCUDAfingerprint recognitionforensicsGPUheterogeneous computing |
| spellingShingle | Andres J. Sanchez-Fernandez Luis F. Romero Daniel Peralta Miguel Angel Medina-Perez Yvan Saeys Francisco Herrera Siham Tabik Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems IEEE Access Biometrics CUDA fingerprint recognition forensics GPU heterogeneous computing |
| title | Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems |
| title_full | Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems |
| title_fullStr | Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems |
| title_full_unstemmed | Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems |
| title_short | Asynchronous Processing for Latent Fingerprint Identification on Heterogeneous CPU-GPU Systems |
| title_sort | asynchronous processing for latent fingerprint identification on heterogeneous cpu gpu systems |
| topic | Biometrics CUDA fingerprint recognition forensics GPU heterogeneous computing |
| url | https://ieeexplore.ieee.org/document/9127429/ |
| work_keys_str_mv | AT andresjsanchezfernandez asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT luisfromero asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT danielperalta asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT miguelangelmedinaperez asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT yvansaeys asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT franciscoherrera asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems AT sihamtabik asynchronousprocessingforlatentfingerprintidentificationonheterogeneouscpugpusystems |