Deteksi AI-Generated Image Menggunakan Convolutional Neural Network

Fikriawan, Akbar (2024) Deteksi AI-Generated Image Menggunakan Convolutional Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003201064-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Pada dua tahun terakhir bidang Generative AI mengalami perkembangan yang sangat pesat terutama dalam bidang Image dan Text Generation. Perkembangan yang sangat pesat tersebut, diiringi dengan kesulitan dalam pendeteksian citra AI-Generated menyebabkan ketakutan dan keresahan pada kalangan luas. Maka dari itu diperlukan dengan cepat sebuah model yang dapat melakukan prediksi antara AI-Generated Image dan Human-Generated Image. Pada penelitian ini dilakukan penyusunan model Convolution Neural Network (CNN) menggunakan transfer learning dari model VGG19 dan MobileNetV3. Didapatkan model terbaik yaitu VGG19 berdasarkan metrik akurasi dan AUC tertinggi dari kedua model. Penelitian ini diharapkan dapat membantu dalam mendeteksi dan mengklasifikasi citra asli dan citra AI.
============================================================
In the last two years, the field of Generative AI has developed very rapidly, especially in the field of Image and Text Generation. This rapid development, accompanied with the difficulties in AI-Generated image detection, has caused fear and unrest in the wider community. Therefore, it is necessary to quickly develop a model that can predict between AI-Generated Image and Human-Generated Image. A Convolution Neural Network (CNN) model was developed using transfer learning from the VGG19 and MobileNetV3 models. The best model is VGG19 with the model achieved the highest in both accuracy and AUC compared to MobileNetV3. This research is hoped to help in detecting and classifying original images and AI images.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	convolutional neural network, ai-generated image, deep learning, generative ai, image generation
Subjects:	Q Science Q Science > Q Science (General) > Q337.5 Pattern recognition systems Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Akbar Fikriawan
Date Deposited:	08 Aug 2024 02:35
Last Modified:	08 Aug 2024 02:35
URI:	http://repository.its.ac.id/id/eprint/114955

Actions (login required)

View Item