CPU and GPU palyginimas vykdant sablonu atitikties algoritma

Image processing, computer vision or other complicated optical information processing algorithms require large resources. It is often desired to execute algorithms in real time. It is hard to fulfill such requirements with single CPU processor. NVidia proposed CUDA technology enables programmer to u...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Science future of Lithuania Ročník 6; číslo 2; s. 129
Hlavní autoři: Borcovas, Evaldas, Daunys, Gintautas
Médium: Journal Article
Jazyk:angličtina
Vydáno: Vilnius Gediminas Technical University 01.04.2014
Témata:
ISSN:2029-2341
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Image processing, computer vision or other complicated optical information processing algorithms require large resources. It is often desired to execute algorithms in real time. It is hard to fulfill such requirements with single CPU processor. NVidia proposed CUDA technology enables programmer to use the GPU resources in the computer. Current research was made with Intel Pentium Dual-Core T4500 2.3 GHz processor with 4 GB RAM DDR3 (CPU I), NVidia GeForce GT320M CUDA compliable graphics card (GPU I) and Intel Core I5-2500K 3.3 GHz processor with 4 GB RAM DDR3 (CPU II), NVidia GeForce GTX 560 CUDA compatible graphic card (GPU II). Additional libraries as OpenCV 2.1 and OpenCV 2.4.0 CUDA compliable were used for the testing. Main test were made with standard function MatchTemplate from the OpenCV libraries. The algorithm uses a main image and a template. An influence of these factors was tested. Main image and template have been resized and the algorithm computing time and performance in Gtpix/s have been measured. According to the information obtained from the research GPU computing using the hardware mentioned earlier is till 24 times faster when it is processing a big amount of information. When the images are small the performance of CPU and GPU are not significantly different. The choice of the template size makes influence on calculating with CPU. Difference in the computing time between the GPUs can be explained by the number of cores which they have. Keywords: image processing, GPGPU, template matching, CUDA. Vaizdu apdorojimas, kompiuterine rega ir kiti sudetingi algoritmai, apdorojantys optine informacija, naudoja didelius skaiciavimo isteklius. Daznai siuos algoritmus reikia realizuoti realiuoju laiku. Si uzdavini isspresti naudojant tik vieno CPU (angl. Central processing unit) pajegumus yra sudetinga. nVidia pasiulyta CUDA (angl. Compute unified device architecture) technologija leidzia panaudoti GPU (angl. Graphic processing unit) isteklius. Tyrimui atlikti buvo pasirinkti du skirtingi CPU: Intel Pentium Dual-Core T4500 ir Intel Core I5 2500K, bei GPU: nVidia GeForce GT320M ir NVidia GeForce 560. Tyrime buvo panaudotos vaizdu apdorojimo bibliotekos: OpenCV 2.1 ir OpenCV 2.4. Tyrimui buvo pasirinktas sablonu atitikties algoritmas. Algoritmui realizuoti reikalingas analizuojamas vaizdas ir ieskomo objekto vaizdo sablonas. Tyrimo metu buvo keiciamas vaizdo ir sablono dydis bei stebima, kaip tai veikia algoritmo vykdymo trukme ir vykdomu operaciju skaiciu per sekunde. Is gautu rezultatu galima teigti, kad apdorojant dideli duomenu kieki GPU realizuoja algoritma iki 24 kartu greiciau nei tik CPU. Dirbant su nedideliu duomenu kiekiu, skirtumas tarp CPU ir GPU yra minimalus. Lyginant skaiciavimus dviejuose GPU, pastebeta, kad skaiciavimu sparta yra tiesiogiai proporcinga GPU turimu branduoliu kiekiui. Musu tyrimo atveju spartesniame GPU ju buvo 16 kartu daugiau, tad ir skaiciavimai vyko 16 kartu sparciau. Reiksminiai zodziai: vaizdu apdorojimas, bendrosios paskirties GPU, sablonu atitiktis, CUDA technologija.
AbstractList Image processing, computer vision or other complicated optical information processing algorithms require large resources. It is often desired to execute algorithms in real time. It is hard to fulfill such requirements with single CPU processor. NVidia proposed CUDA technology enables programmer to use the GPU resources in the computer. Current research was made with Intel Pentium Dual-Core T4500 2.3 GHz processor with 4 GB RAM DDR3 (CPU I), NVidia GeForce GT320M CUDA compliable graphics card (GPU I) and Intel Core I5-2500K 3.3 GHz processor with 4 GB RAM DDR3 (CPU II), NVidia GeForce GTX 560 CUDA compatible graphic card (GPU II). Additional libraries as OpenCV 2.1 and OpenCV 2.4.0 CUDA compliable were used for the testing. Main test were made with standard function MatchTemplate from the OpenCV libraries. The algorithm uses a main image and a template. An influence of these factors was tested. Main image and template have been resized and the algorithm computing time and performance in Gtpix/s have been measured. According to the information obtained from the research GPU computing using the hardware mentioned earlier is till 24 times faster when it is processing a big amount of information. When the images are small the performance of CPU and GPU are not significantly different. The choice of the template size makes influence on calculating with CPU. Difference in the computing time between the GPUs can be explained by the number of cores which they have.
Image processing, computer vision or other complicated optical information processing algorithms require large resources. It is often desired to execute algorithms in real time. It is hard to fulfill such requirements with single CPU processor. NVidia proposed CUDA technology enables programmer to use the GPU resources in the computer. Current research was made with Intel Pentium Dual-Core T4500 2.3 GHz processor with 4 GB RAM DDR3 (CPU I), NVidia GeForce GT320M CUDA compliable graphics card (GPU I) and Intel Core I5-2500K 3.3 GHz processor with 4 GB RAM DDR3 (CPU II), NVidia GeForce GTX 560 CUDA compatible graphic card (GPU II). Additional libraries as OpenCV 2.1 and OpenCV 2.4.0 CUDA compliable were used for the testing. Main test were made with standard function MatchTemplate from the OpenCV libraries. The algorithm uses a main image and a template. An influence of these factors was tested. Main image and template have been resized and the algorithm computing time and performance in Gtpix/s have been measured. According to the information obtained from the research GPU computing using the hardware mentioned earlier is till 24 times faster when it is processing a big amount of information. When the images are small the performance of CPU and GPU are not significantly different. The choice of the template size makes influence on calculating with CPU. Difference in the computing time between the GPUs can be explained by the number of cores which they have. Keywords: image processing, GPGPU, template matching, CUDA. Vaizdu apdorojimas, kompiuterine rega ir kiti sudetingi algoritmai, apdorojantys optine informacija, naudoja didelius skaiciavimo isteklius. Daznai siuos algoritmus reikia realizuoti realiuoju laiku. Si uzdavini isspresti naudojant tik vieno CPU (angl. Central processing unit) pajegumus yra sudetinga. nVidia pasiulyta CUDA (angl. Compute unified device architecture) technologija leidzia panaudoti GPU (angl. Graphic processing unit) isteklius. Tyrimui atlikti buvo pasirinkti du skirtingi CPU: Intel Pentium Dual-Core T4500 ir Intel Core I5 2500K, bei GPU: nVidia GeForce GT320M ir NVidia GeForce 560. Tyrime buvo panaudotos vaizdu apdorojimo bibliotekos: OpenCV 2.1 ir OpenCV 2.4. Tyrimui buvo pasirinktas sablonu atitikties algoritmas. Algoritmui realizuoti reikalingas analizuojamas vaizdas ir ieskomo objekto vaizdo sablonas. Tyrimo metu buvo keiciamas vaizdo ir sablono dydis bei stebima, kaip tai veikia algoritmo vykdymo trukme ir vykdomu operaciju skaiciu per sekunde. Is gautu rezultatu galima teigti, kad apdorojant dideli duomenu kieki GPU realizuoja algoritma iki 24 kartu greiciau nei tik CPU. Dirbant su nedideliu duomenu kiekiu, skirtumas tarp CPU ir GPU yra minimalus. Lyginant skaiciavimus dviejuose GPU, pastebeta, kad skaiciavimu sparta yra tiesiogiai proporcinga GPU turimu branduoliu kiekiui. Musu tyrimo atveju spartesniame GPU ju buvo 16 kartu daugiau, tad ir skaiciavimai vyko 16 kartu sparciau. Reiksminiai zodziai: vaizdu apdorojimas, bendrosios paskirties GPU, sablonu atitiktis, CUDA technologija.
Audience Academic
Author Daunys, Gintautas
Borcovas, Evaldas
Author_xml – sequence: 1
  fullname: Borcovas, Evaldas
– sequence: 2
  fullname: Daunys, Gintautas
BookMark eNptkE9LAzEQxXOoYK09-QUWPO-aZHb2D3gpi1ahoId6LpNNdonNZqWJQr-9AT148M3hweM3M_Cu2MLP3jB2I3gBTVndTY4KyUVZiGrBlpLLNpdQiku2DuGdJyGXvMElu-9e3zLyOtsm_yB3Hq23E4Xs63zU5GMWSLnZf2YUbbTHaE3IyI3zycaJrtnFQC6Y9a-v2P7xYd895buX7XO32eVjVUPeqLYBRTgopLIl3qf3UrXIjSFChIFKUFjXQlfCoAYkbBDLVgmNmjcSVuz25-xIzhysH-Z4on6yoT9soK5aAI6QqOIfKo02k-1TPYNN-Z-FbzN-WEk
ContentType Journal Article
Copyright COPYRIGHT 2014 Vilnius Gediminas Technical University
Copyright_xml – notice: COPYRIGHT 2014 Vilnius Gediminas Technical University
DOI 10.3846/mla.2014.16
DatabaseTitleList

DeliveryMethod fulltext_linktorsrc
Discipline Sciences (General)
ExternalDocumentID A376933053
GeographicLocations Lithuania
GeographicLocations_xml – name: Lithuania
GroupedDBID 5VS
7X2
7XC
8CJ
8FE
8FG
8FH
8G5
ABDBF
ABJCF
ABUWG
ACIWK
ACPRK
ACUHS
ADBBV
AEUYN
AFFHD
AFKRA
AFRAH
ALMA_UNASSIGNED_HOLDINGS
ARAPS
ATCPS
AZQEC
BBNVY
BCNDV
BENPR
BGLVJ
BHPHI
BKSAR
BPHCQ
BYOGL
CCPQU
D1I
D1J
D1K
DWQXO
EOJEC
ESX
GNUQQ
GROUPED_DOAJ
GUQSH
H13
HCIFZ
IAO
IEA
IGS
IPNFZ
ITC
K6-
K6V
K7-
KB.
KQ8
L6V
LK5
LK8
M0K
M2O
M7P
M7R
M7S
OBODZ
P62
PADUT
PATMY
PCBAR
PDBOC
PHGZM
PHGZT
PIMPY
PQGLB
PQQKQ
PROAC
PTHSS
PYCSY
RIG
ID FETCH-LOGICAL-g673-8b983ba5fb5a49a0c0002b950eeaa553fa43b5771d61e5d35a585549b1d5d0823
ISSN 2029-2341
IngestDate Tue Nov 11 10:36:57 EST 2025
Tue Nov 04 17:50:01 EST 2025
IsPeerReviewed true
IsScholarly true
Issue 2
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-g673-8b983ba5fb5a49a0c0002b950eeaa553fa43b5771d61e5d35a585549b1d5d0823
ParticipantIDs gale_infotracmisc_A376933053
gale_infotracacademiconefile_A376933053
PublicationCentury 2000
PublicationDate 20140401
PublicationDateYYYYMMDD 2014-04-01
PublicationDate_xml – month: 04
  year: 2014
  text: 20140401
  day: 01
PublicationDecade 2010
PublicationTitle Science future of Lithuania
PublicationYear 2014
Publisher Vilnius Gediminas Technical University
Publisher_xml – name: Vilnius Gediminas Technical University
SSID ssj0000502085
Score 1.8669158
Snippet Image processing, computer vision or other complicated optical information processing algorithms require large resources. It is often desired to execute...
SourceID gale
SourceType Aggregation Database
StartPage 129
SubjectTerms Analysis
Evaluation
Image processing
Machine vision
Title CPU and GPU palyginimas vykdant sablonu atitikties algoritma
Volume 6
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals (WRLC)
  issn: 2029-2341
  databaseCode: DOA
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.doaj.org/
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: Directory of Open Access Journals
– providerCode: PRVPQU
  databaseName: Advanced Technologies & Aerospace Database
  issn: 2029-2341
  databaseCode: P5Z
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://search.proquest.com/hightechjournals
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Agricultural Science Database
  issn: 2029-2341
  databaseCode: M0K
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://search.proquest.com/agriculturejournals
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Biological Science Database
  issn: 2029-2341
  databaseCode: M7P
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com/biologicalscijournals
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Computer Science Database (ProQuest)
  issn: 2029-2341
  databaseCode: K7-
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com/compscijour
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Earth, Atmospheric & Aquatic Science Database
  issn: 2029-2341
  databaseCode: PCBAR
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://search.proquest.com/eaasdb
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: East Europe, Central Europe Database (ProQuest)
  issn: 2029-2341
  databaseCode: BYOGL
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://search.proquest.com/eastcentraleurope
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Engineering Database
  issn: 2029-2341
  databaseCode: M7S
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Environmental Science Database
  issn: 2029-2341
  databaseCode: PATMY
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com/environmentalscience
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Materials Science Database
  issn: 2029-2341
  databaseCode: KB.
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com/materialsscijournals
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  issn: 2029-2341
  databaseCode: BENPR
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.proquest.com/central
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Publicly Available Content Database
  issn: 2029-2341
  databaseCode: PIMPY
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: http://search.proquest.com/publiccontent
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: Research Library
  issn: 2029-2341
  databaseCode: M2O
  dateStart: 20090101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://search.proquest.com/pqrl
  omitProxy: false
  ssIdentifier: ssj0000502085
  providerName: ProQuest
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Lb9QwELagcOCCKA9RKJUPSICiQBLHSSxxQRUsEqjqYal6qxzHWaym3iqPqv33zDjeTeipHLgkUR5W4m_y2R6PvyHkLc-FlrpiYSpYFmKWo7CUaR6qupSqyuJMO1Gfk5_50VFxeiqOfa72zqUTyK0trq_F5X-FGs4B2Lh09h_g3hYKJ-AYQIctwA7bOwF_ePzLzQgsYH8pm5uVseZCdsHVzXkF1Rh0smzWdgjwdnOOeqqBbFbr1vSeon1XdfPXj6ojbi7B9L8Hac00hl-3CmNZHaHCN1RylqZ-sKONLIztJUZ-zx0M8TwuBU3ixDTWDF2wgMYUo3O60envTGgKHploK8GAmoSNclYbjs1mppTM-DIe3R23eZwVTsD4okFlqDj9GN9Syx4FexnmcQS-YvfJgyTnAnn5Rx5uXWwRdxlIMb3g5p3GFZpY_KdZ4b45nnUslk_IYz8ioF9GJHfJPW2fkl1f-x1974XBPzwjnwFaCtBSgJbOoKUeWuqhpRO0dAvtc7L89nV5-D302S_CVZazsChFwUrJ65LLVMhIYdtVCh5pLSXnrJYpK3mex_BDaV4xLjlGHIoyrniF06cvyI5dW_2SUBiy1knEal3B2FglODMN_fy6YFkcq0ile-Qdfv4ZItC3Ukm_MgOeRnGws6mi98j-X3cCFanZ5Vd3Lug1eTSZ2j7Z6dtBvyEP1VVvuvbAOUYOHJh_AOYXVak
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=CPU+and+GPU+palyginimas+vykdant+sablonu+atitikties+algoritma&rft.jtitle=Science+future+of+Lithuania&rft.au=Borcovas%2C+Evaldas&rft.au=Daunys%2C+Gintautas&rft.date=2014-04-01&rft.pub=Vilnius+Gediminas+Technical+University&rft.issn=2029-2341&rft.volume=6&rft.issue=2&rft.spage=129&rft_id=info:doi/10.3846%2Fmla.2014.16&rft.externalDocID=A376933053
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2029-2341&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2029-2341&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2029-2341&client=summon