JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI
Vol 10 No 3 (2023): JATISI (Jurnal Teknik Informatika dan Sistem Informasi)

CONVOLUTIONAL NEURAL NETWORK PADA IDENTIFIKASI VARIAN TANAMAN ANGGUR MENGGUNAKAN RESNET-50

Muhammad Ali Nur Hidayat (Unknown)



Article Info

Publish Date
06 Sep 2023

Abstract

Identification of grape plant variants is very important, as an initial step to determine the validity of planted grape seeds. Grape variants can be distinguished after the vine bears fruit, but this method takes 8 months to 2 years from planting depending on the care taken. the second fastest way is to look at the shape of the grape leaves, but this method can only be done by people who are experts in recognizing the variants/types of grapes, while ordinary people will find it difficult to tell the variants of the grapes from the shape of the leaves. This study will use the CNN Restnet-50 method to identify grapevine variants based on grape leaf images. There are 150 grape leaf image data used with 3 grape variants consisting of 50 Nizina leaves, 50 Jupiter leaves and 50 Isabella leaves. grape leaf image data resized in pixels, remove background, data augmentation. The system produces a training accuracy rate of 86% while the best validation accuracy is 91%.

Copyrights © 2023






Journal Info

Abbrev

jatisi

Publisher

Subject

Computer Science & IT

Description

JATISI bekerja sama dengan IndoCEISS dalam pengelolaannya. IndoCEISS merupakan wadah bagi para ilmuwan, praktisi, pendidik, dan penggemar dalam bidang komputer, elektronika, dan instrumentasi yang menaruh minat untuk memajukan bidang tersebut di Indonesia. JATISI diterbitkan 2 kali dalam setahun ...