Perbandingan Model AlexNet dan ResNet dalam Klasifikasi Citra Bunga Memanfaatkan Transfer Learning

Bana Falakhi (Universitas Brawijaya)
Elmira Faustina Achmal (Universitas Brawijaya)
Muhammad Rizaldi (Universitas Brawijaya)
Renata Rizki Rafi' Athallah (Universitas Brawijaya)
Novanto Yudistira (Universitas Brawijaya)

Jurnal Ilmu Komputer dan Agri-Informatika
Vol 9 No 1 (2022)

Image-based automatic flower species classification is an important issue for biologists creating digital flower catalogs. Many studies on flower species recognition have been proposed so far based on traditional image processing routines. Currently, researchers are applying deep learning to various image-based object recognition tasks. In this paper, deep learning based on transfer learning is applied to the classification of flower species. The proposed methoduses AlexNet and ResNet transfer learning models. The Flower102 dataset which has many categories is used in the experimental work. Various experimental results show that each model has achieved 87% and 96% accuracy performance for AlexNet and ResNet. Theresults obtained show that the effectiveness of the ResNet-based model is higher than the AlexNet-based model.