Classification of Clove Leaf Blister Blight Disease Severity Using Pre-trained Model VGG16, InceptionV3, and ResNet

Pramesti, Putri Ayu
Supriyadi, Muhamad Rodhi
Alfin, Muhammad Reza
Noveriza, Rita
Wahyuno, Dono
Manohara, Dyah
Melati
Miftakhurohmah
Warman, Riki
Hardiyanti, Siti
Asnawi

Jurnal Ilmu Komputer dan Informasi
Vol. 17 No. 2 (2024): Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Informatio

Clove is one of the precious plants produced in Indonesia. Clove has many benefits for humans, but clove cultivation often experiences problems due to disease attacks, including Leaf Blister Blight Disease(CDC). The handling of CDC disease is carried out based on the severity of the symptoms that can be seen on the affected leaves. This research was conducted to obtain a CDC disease classification model, so appropriate treatment can be carried out. This study used the pre-trained VGG16, InceptionV3, and ResNet models for classification. VGG16 got the highest average accuracy of 96.7%. Aside from that, k-fold cross validation improved the model's accuracy.