Recent Advances in Computer Vision Applications Using Parallel Processing

126 Pages·2023·2.996 MB·English

by  Khalid M. Hosny, Ahmad Salah

Studies in Computational Intelligence 1073
Khalid M. Hosny
Ahmad Salah   Editors
Recent Advances
in Computer Vision
Applications Using
Parallel Processing
Studies in Computational Intelligence
Volume 1073
SeriesEditor
JanuszKacprzyk,PolishAcademyofSciences,Warsaw,Poland
The series “Studies in Computational Intelligence” (SCI) publishes
new developments and advances in the various areas of computational
intelligence—quickly and with a high quality. The intent is to cover the theory,
applications, and design methods of computational intelligence, as embedded in
the fields of engineering, computer science, physics and life sciences, as well as
themethodologiesbehindthem.Theseriescontainsmonographs,lecturenotesand
editedvolumesincomputationalintelligencespanningtheareasofneuralnetworks,
connectionist systems, genetic algorithms, evolutionary computation, artificial
intelligence, cellular automata, self-organizing systems, soft computing, fuzzy
systems,andhybridintelligentsystems.Ofparticularvaluetoboththecontributors
and the readership are the short publication timeframe and the world-wide
distribution,whichenablebothwideandrapiddisseminationofresearchoutput.
IndexedbySCOPUS,DBLP,WTIFrankfurteG,zbMATH,SCImago.
AllbookspublishedintheseriesaresubmittedforconsiderationinWebofScience.
·
Khalid M. Hosny Ahmad Salah
Editors
Recent Advances
in Computer Vision
Applications Using Parallel
Processing
Editors
KhalidM.Hosny AhmadSalah
DepartmentofInformationTechnology CollegeofComputingandInformation
ZagazigUniversity Sciences
Sharkia,Egypt UniversityofTechnologyandApplied
Sciences
Ibri,Oman
ISSN 1860-949X ISSN 1860-9503 (electronic)
StudiesinComputationalIntelligence
ISBN 978-3-031-18734-6 ISBN 978-3-031-18735-3 (eBook)
https://doi.org/10.1007/978-3-031-18735-3
©TheEditor(s)(ifapplicable)andTheAuthor(s),underexclusivelicensetoSpringerNature
SwitzerlandAG2023
Thisworkissubjecttocopyright.AllrightsaresolelyandexclusivelylicensedbythePublisher,whether
thewholeorpartofthematerialisconcerned,specificallytherightsoftranslation,reprinting,reuse
ofillustrations,recitation,broadcasting,reproductiononmicrofilmsorinanyotherphysicalway,and
transmissionorinformationstorageandretrieval,electronicadaptation,computersoftware,orbysimilar
ordissimilarmethodologynowknownorhereafterdeveloped.
Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication
doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant
protectivelawsandregulationsandthereforefreeforgeneraluse.
Thepublisher,theauthors,andtheeditorsaresafetoassumethattheadviceandinformationinthisbook
arebelievedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsor
theeditorsgiveawarranty,expressedorimplied,withrespecttothematerialcontainedhereinorforany
errorsoromissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictional
claimsinpublishedmapsandinstitutionalaffiliations.
ThisSpringerimprintispublishedbytheregisteredcompanySpringerNatureSwitzerlandAG
Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland
Preface
Computervisionisonefieldthatisconsideredcompute-intensive.Thisisbecause
the input can be an image or a video. As the input is of large size, then the
computing/processingtimeishugeaswell.Inaddition,deeplearningmethodsare
heavilyusedinthefieldofcomputervision.Thus,utilizingtheparallelarchitecture
for improving the runtime of the computer vision methods is of great interest and
benefit. Of note, the inputs of the computer vision methods are parallel-friendly.
Forinstance,oneimageconsistsofasetofpixels.Thosepixelscanbedividedinto
groupsbasedonthenumberofavailableparallelresourcesandtheneachgroupof
pixelsisprocessedusingonecomputationalresource(e.g.,CPU).Thus,thegroups
of pixels are processed in parallel. Similarly, the video input can be divided into
frames,whereeachcomputationalresource(e.g.,CPUorGPU)handlesanumber
offrames.
The book provides the readers with a comprehensive overview of principles,
methodologies, and recent advances in utilizing parallel architectures for the sake
ofspeedingupcomputervisionmethodsandalgorithms.Thiseditedbookcontains
six chapters. These chapters feature original and previously unpublished works of
renownedscholarsfrommultiplenationsthataddressvariousaspectsofparallelizing
computer vision methods. Each chapter begins with an introduction that covers
contemporary methods, discusses results, and identifies difficulties and potential
optionsforthefuture.AGenericMulticoreCPUParallelImplementationforFrac-
tional Order Digital Image Moments. Additionally, each chapter provides a list of
referencesforfurtherresearch.Researchers,engineers,ITexperts,developers,post-
graduate students, and senior undergraduate students majoring in computer vision
orhigh-performancecomputingwillfindthebooktobeausefulreference.
Abriefoverviewofthecontentsofthebookisasfollows:Theauthorsinthefirst
chapterproposedagenericmethodforparallelizinganyfractionalordermomenton
theparallelarchitectureofmulti-coreCPUsinthepaperentitled“AGenericMulti-
core CPU Parallel Implementation for Fractional Order Digital Image Moments”.
They proposed accelerating two well-known algorithms, namely, fractional polar
cosinetransform(Fr-PCT)andfractionalpolarsinetransform(FrPST)onCPUcore
with 16 cores with the help of the OpenMP APIs. They managed to speed up the
v
vi Preface
two algorithms by 15.9 times. In second chapter, the authors reviewed the recent
advancesinthetopicofroadinspectioninthepaperentitled“Computer-AidedRoad
Inspection: Systems and Algorithms”. The second chapter first compares the five
mostcommonroaddamagetypes.Then,2D/3Droadimagingsystemsarediscussed.
Finally, state-of-the-art machine vision and intelligence-based road damage detec-
tion algorithms are introduced. Third chapter is entitled “Computer Stereo Vision
forAutonomous Driving:Theory and Algorithms”.Inthischapter, toenhance the
trade-offbetweenspeedandaccuracy,acomputerstereovisionalgorithmforhard-
warewithlimitedresourcesisdevelopedbytheauthors.Thethirdchapterintroduces
thehardwareandsoftwarecomponentsoftheautonomousvehiclesystem.Following
that,theauthorsgothroughfourtasksforautonomouscarstoperformintermsof
perception,including(1)visualfeatureidentification,description,andmatching;(2)
3Dinformationacquisition;(3)objectdetectionandrecognition;and(4)semantic
picturesegmentation.Theconceptsofparallelcomputingonmulti-threadingCPU
and GPU are then described in depth. In Fourth chapter, the authors reviewed the
recent advancements in the field of using GPU to accelerate visual object trackers
in the paper entitled “A Survey on GPU-Based Visual Trackers”. In this chapter,
Single Object Tracking (SOT) and Multiple Object Tracking (MOT) are catego-
rized, summarized, and analyzed. Then, the authors discussed parallel computing
applications for object tracking methods, and numerous techniques for evaluating
the performance of parallel algorithms on parallel architectures. In Fifth chapter,
theauthorsproposedacceleratingthecopy-moveforgerydetectionindigitalimages
in the paper entitled “Accelerating the Process of Copy-Move Forgery Detection
Using Multi-core CPUs Parallel Architecture”. Several copy-move forgery detec-
tion methods are introduced, but the fundamental problem with them is the huge
runtime required to find the final region that was forged. This is because several
intricatecomputationaloperationsmustberepeated.Usingparallelarchitecturesis
oneofthesolutionsthathavebeenproposedtoovercomethisissue.Althoughthere
hasbeenminimalstudygiveninthisdirection,itisstillinitsearlystages.Theuse
ofparallelarchitecturesisexposedinthisarticle.Theauthorsconsidereddifferent
parallel methods to speed up the copy-move forgery detection process in order to
addresstheissueofmassiveruntime.Inthiscontext,anovelMulti-coreCPU-based
ParallelCopy-MoveImageForgeryDetection(PCMIFD)systemisproposedbythe
authors.Theproposedsystemhasundergone testingonan8-coremulti-coreCPU
parallelarchitecture;theobtainedresultsshowthattheproposedsystemmanagedto
speedupthesequentialcopy-forgeryalgorithmby7.4times.Finally,inSixthchapter,
theauthorsproposedreviewingtherecentresearchworksconductedusingRaspberry
Piwithmulti-coreCPUembeddedsystemsoraclusterofRaspberryPitospeedup
thecomputervisiontechniques.Thischapterisentitled“ParallelImageProcessing
ApplicationsUsingRaspberryPi”.InsteadofusingnormalCPUs,aclusteroflow-
power embedded processors can reduce the electricity consumption. Because the
portableclustermaybesetuptokeeprunningevenifsomeofitsnodesfail,itwill
beadvantageoustoemployaRaspberryPiclusterforimageprocessingapplications
thattakealongtimetocomplete.Theauthorsofthispapergiveageneraloverview
Preface vii
of how the Raspberry Pi is used in various parallel image processing applications
acrossseveralindustries.
Sharkia,Egypt KhalidM.Hosny
Ibri,Oman AhmadSalah
About This Book
Parallel architectures have become commonplace. These architectures helped
researcherstoaddresspreviouslyinfeasibleproblemsintermsofruntimeduetothe
greatimprovementintheelapsedruntime.Forinstance,thesuccessofthetraining
deepNeuralNetworks(NNs)isowedtotheGraphicalProcessingUnits(GPUs);this
ideaoftrainingmassivelydeepNNswastherefordecades,butitbecametrueafter
successfullyadvancingGPUs.
ix
Contents
AGenericMulticoreCPUParallelImplementationforFractional
OrderDigitalImageMoments ...................................... 1
AhmadSalah,KhalidM.Hosny,andAmrM.Abdeltif
Computer-AidedRoadInspection:SystemsandAlgorithms ........... 13
RuiFan,SicenGuo,LiWang,andMohammudJunaidBocus
Computer Stereo Vision for Autonomous Driving: Theory
andAlgorithms ................................................... 41
RuiFan,LiWang,MohammudJunaidBocus,andIoannisPitas
ASurveyonGPU-BasedVisualTrackers ............................ 71
IslamMohamed,IbrahimElhenawy,andAhmadSalah
AcceleratingtheProcessofCopy-MoveForgeryDetectionUsing
Multi-coreCPUsParallelArchitecture .............................. 87
HanaaM.Hamza,KhalidM.Hosny,andAhmadSalah
ParallelImageProcessingApplicationsUsingRaspberryPi ........... 107
KhalidM.Hosny,AhmadSalah,andAmalMagdi
xi