RANCANG BANGUN APLIKASI UJI KEMIRIPAN GAMBAR AI GENERATIVE DAN GAMBAR BUATAN TANGAN MENGGUNAKAN METODE DEEP LEARNING

Rifqi Alfaesta Prawiratama (Universitas Muhammadiyah Sidoarjo)
Sumarno Sumarno (Universitas Muhammadiyah Sidoarjo)
Irwan Alnarus Kautsar (Universitas Muhammadiyah Sidoarjo)

Jurnal Tekinkom (Teknik Informasi dan Komputer)
Vol 7 No 1 (2024)

This research discusses the development of an application to test the similarity between AI Generative images and handmade images using deep learning methods. Artificial Intelligence (AI) technology has been applied to generative art through deep learning algorithms; however, there are still challenges related to copyright and originality of AI Generative art. The aim of this research is to develop an efficient model for classifying AI Generative art and handmade art. The classification model uses a Transformer approach, specifically exploiting the BEiT architecture, which shows highly satisfactory results in image classification tests. The high F1 score in each test reflects a good balance between precision and recall. The Transformer model outperforms previous methods using Convolutional Neural Network (CNN) and VGG16 models. It is expected that this model will be able to classify art more efficiently, assist in the detection of misuse, and mitigate legal risks related to copyright.