أعضاء هيئة التدريس - جامعة بنها
Automatic Gun Detection Approach for Video Surveillance
International Journal of Sociotechnology and Knowledge Development • 2020
معلومات البحث
المؤلفون
Mai K. Galab; Ahmed Taha; Hala H. Zayed
الكلمات المفتاحية
Closed Circuit Television, Convolutional Neural Network, Deep Neural Network, Transfer Learning, Video
Surveillance Systems
المجلة العلمية
International Journal of Sociotechnology and Knowledge Development
الناشر
IGI Global
المجلد
Not Available
العدد
Not Available
الصفحات
Not Available
publication.type
International
رابط البحث
Not Available
المواد المرفقة
Not Available
الملخص
Theimmensecrimeratesresultingfromusingpistolshaveledgovernmentstoseeksolutionstodeal
withsuchterroristincidents.Theseincidentshaveanegativeimpactonpublicsecurityandcause
panicamongcitizens.Fromthispoint,facingapandemicofweaponviolencehasbecomeanimportant
researchtopic.Onewaytoreducethiskindofviolenceistopreventitviaremotedetectionandtogive
anappropriateresponseinashorttime.Videosurveillanceistheprocessofmonitoringthebehavior
ofpeopleandobjects.Surveillancesystemscanbeemployedinsecurityapplicationsaslegalevidence.
Moreover,itisusedwidelyinsuspiciousactivitydetectionapplications.Intelligentvideosurveillance
systems(IVSSs)aretheuseofautomaticvideoanalyticstoenhancetheeffectivenessoftraditional
surveillancesystems.WiththerapiddevelopmentinDeepLearning(DL),itisnowwidelyusedto
addresstheproblemsexistingintraditionaldetectiontechniques.Inthisarticle,anapproachtodetect
pistolsandgunsinvideosurveillancesystemsisproposed.Thepresentedapproachdoesnotneed
anyinvasivetoolsintheweapondetectionprocess.ItusesDLintheclassificationandthedetection
processes.TheproposedapproachenhancestheobtainedresultsbyapplyingTransferLearning(TL).
ItemploystwodifferentDLtechniques:AlexNetandGoogLeNet.Experimentalresultsverifythe
adaptabilityofdetectingdifferenttypesofpistolsandguns.Theexperimentswereconductedona
benchmarkgundatabasecalledInternetMovieFirearmsDatabase(IMFDB).Theresultsobtained
suggestthattheproposedapproachispromisingandoutperformsitscounterparts.
withsuchterroristincidents.Theseincidentshaveanegativeimpactonpublicsecurityandcause
panicamongcitizens.Fromthispoint,facingapandemicofweaponviolencehasbecomeanimportant
researchtopic.Onewaytoreducethiskindofviolenceistopreventitviaremotedetectionandtogive
anappropriateresponseinashorttime.Videosurveillanceistheprocessofmonitoringthebehavior
ofpeopleandobjects.Surveillancesystemscanbeemployedinsecurityapplicationsaslegalevidence.
Moreover,itisusedwidelyinsuspiciousactivitydetectionapplications.Intelligentvideosurveillance
systems(IVSSs)aretheuseofautomaticvideoanalyticstoenhancetheeffectivenessoftraditional
surveillancesystems.WiththerapiddevelopmentinDeepLearning(DL),itisnowwidelyusedto
addresstheproblemsexistingintraditionaldetectiontechniques.Inthisarticle,anapproachtodetect
pistolsandgunsinvideosurveillancesystemsisproposed.Thepresentedapproachdoesnotneed
anyinvasivetoolsintheweapondetectionprocess.ItusesDLintheclassificationandthedetection
processes.TheproposedapproachenhancestheobtainedresultsbyapplyingTransferLearning(TL).
ItemploystwodifferentDLtechniques:AlexNetandGoogLeNet.Experimentalresultsverifythe
adaptabilityofdetectingdifferenttypesofpistolsandguns.Theexperimentswereconductedona
benchmarkgundatabasecalledInternetMovieFirearmsDatabase(IMFDB).Theresultsobtained
suggestthattheproposedapproachispromisingandoutperformsitscounterparts.