?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=PERBANDINGAN+METODE+YOLOv4-MobileNetV3+DAN+YOLOv7%0D%0APADA+DETEKSI+DAN+KLASIFIKASI+PLAT+NOMOR+KENDARAAN&rft.creator=AGES%2C+MAHESA&rft.subject=000+Ilmu+komputer%2C+informasi+dan+pekerjaan+umum&rft.subject=500+ilmu+pengetahuan+alam+dan+matematika&rft.description=Plat+Kendaraan+berfungsi+sebagai+tanda+pengenal+resmi+untuk+kendaraan+bermotor.%0D%0ADi+Indonesia%2C+plat+kendaraan+memiliki+empat+kategori+warna+berbeda%2C+yaitu+putih%0D%0Adengan+tulisan+hitam%2C+kuning+dengan+tulisan+hitam%2C+merah+dengan+tulisan+putih%2C+dan%0D%0Ahijau+dengan+tulisan+hitam.+Regulasi+baru+mengubah+warna+plat+dari+plat+hitam%0D%0Adengan+tulisan+putih+menjadi+plat+putih+dengan+tulisan+hitam+untuk+menghindari%0D%0Akesalahan+deteksi+yang+dilakukan+sistem+E-tilang+ketika+mendeteksi+plat+hitam%0D%0Adengan+tulisan+putih.+Penelitian+ini+bertujuan+untuk+mendeteksi+dan%0D%0Amengklasifikasikan+empat+warna+plat+kendaraan%2C+yaitu+hitam%2C+putih%2C+kuning%2C+dan%0D%0Amerah%2C+serta+kendaraan+non+plat+menggunakan+metode+YOLOv4+dengan%0D%0AMobileNetV3+sebagai+fitur+ekstraktor+dan+metode+YOLOv7.+Kedua+metode+akan%0D%0Adilakukan+evaluasi+performa+untuk+melakukan+perbandingan+kinerja+dalam+tugas%0D%0Adeteksi+dan+klasifikasi+objek+plat+kendaraan.+Hasil+penelitian+dengan+menggunakan%0D%0Adata+uji+berupa+video+digital%2C+menunjukkan+bahwa+YOLOv7+unggul+dalam+performa%0D%0Adeteksi+dengan+rata-rata+precision+95.24%25%2C+recall+94.68%25%2C+dan+f1-score+94.95%25%2C%0D%0Asementara+YOLOv4-MobileNetV3+memiliki+rata-rata+precision+93.88%25%2C+recall%0D%0A93.45%25%2C+dan+f1-score+93.66%25.+Namun%2C+pada+evaluasi+running+time+YOLOv4-%0D%0AMobileNetV3+lebih+unggul+dengan+rata-rata+waktu+komputasi+mencapai+FPS+51.52%2C%0D%0Adibandingkan+dengan+YOLOv7+yang+hanya+mencapai+rata-rata+FPS+34.44.%0D%0APenelitian+ini+menunjukkan+bahwa+metode+YOLOv7+lebih+unggul+dalam+hal+akurasi%0D%0Adeteksi+dan+klasifikasi%2C+sementara+YOLOv4-MobileNetV3+lebih+efisien+dalam%0D%0Awaktu+komputasi.%0D%0A%0D%0AKata+Kunci%3A+Deep+Learning%2C+YOLOv4%2C+YOLOv7%2C+MobileNetV3%2C+Deteksi+Objek%2C%0D%0AKlasifikasi%2C+Plat+Kendaraan.%0D%0A%0D%0ALicense+plates+serve+as+official+identification+markers+for+motor+vehicles.+In%0D%0AIndonesia%2C+license+plates+have+four+different+color+categories%3A+white+with+black+text%2C%0D%0Ayellow+with+black+text%2C+red+with+white+text%2C+and+green+with+black+text.+A+new%0D%0Aregulation+has+changed+the+color+of+license+plates+from+black+with+white+text+to%0D%0Awhite+with+black+text+to+avoid+detection+errors+by+the+E-tilang+system%2C+which%0D%0Amisinterpreted+black+plates+with+white+text.+This+research+aims+to+detect+and+classify%0D%0Afour+license+plate+colors%3A+black%2C+white%2C+yellow%2C+and+red%2C+as+well+as+non-plate%0D%0Avehicles%2C+using+the+YOLOv4+method+with+MobileNetV3+as+the+feature+extractor+and%0D%0Athe+YOLOv7+method.+Both+methods+are+evaluated+to+compare+their+performance+in%0D%0Adetecting+and+classifying+license+plates.+The+results+of+this+research%2C+using+test+data%0D%0Ain+the+form+of+digital+videos%2C+indicate+that+YOLOv7+excels+in+detection+performance%0D%0Awith+an+average+precision+of+95.24%25%2C+recall+of+94.68%25%2C+and+an+F1-score+of+94.95%25.%0D%0AIn+contrast%2C+YOLOv4-MobileNetV3+achieves+an+average+precision+of+93.88%25%2C+recall%0D%0Aof+93.45%25%2C+and+an+F1-score+of+93.66%25.+However%2C+in+terms+of+running+time%0D%0Aevaluation%2C+YOLOv4-MobileNetV3+is+superior+with+an+average+computational+time%0D%0Aof+51.52+FPS%2C+compared+to+YOLOv7%2C+which+reaches+only+an+average+of+34.44+FPS.%0D%0AThis+study+demonstrates+that+the+YOLOv7+method+is+better+in+terms+of+detection+and%0D%0Aclassification+accuracy%2C+while+YOLOv4-MobileNetV3+is+more+efficient+in%0D%0Acomputational+time.%0D%0A%0D%0AKata+Kunci%3A+Deep+Learning%2C+YOLOv4%2C+YOLOv7%2C+MobileNetV3%2C+Object+Detection%2C%0D%0AClassification%2C+License+Plate.&rft.publisher=FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM&rft.date=2024-07-19&rft.type=Skripsi&rft.type=NonPeerReviewed&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F86758%2F1%2FFile%2520ABSTRAK%2520-%2520Ages%2520Mahesa.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F86758%2F2%2FFILE%2520FULL%2520SKRIPSI%2520-%2520Ages%2520Mahesa.pdf&rft.format=text&rft.identifier=http%3A%2F%2Fdigilib.unila.ac.id%2F86758%2F3%2FFILE%2520FULL%2520SKRIPSI%2520TANPA%2520PEMBAHASAN%2520-%2520Ages%2520Mahesa.pdf&rft.identifier=++AGES%2C+MAHESA++(2024)+PERBANDINGAN+METODE+YOLOv4-MobileNetV3+DAN+YOLOv7+PADA+DETEKSI+DAN+KLASIFIKASI+PLAT+NOMOR+KENDARAAN.++FAKULTAS+MATEMATIKA+DAN+ILMU+PENGETAHUAN+ALAM%2C+UNIVERSITAS+LAMPUNG.+++++&rft.relation=http%3A%2F%2Fdigilib.unila.ac.id%2F86758%2F