DETECTION OF AI GENERATED IMAGES USING GATED EXPERT CONVOLUTIONAL NEURAL NETWORKS

ANGIREKULA SINDU SRI, GOLLA ARUNA, GOMARI KUSHI, GORRE PRAVALLIKA, MRS. D. GEETHA

doi:10.62643/

Authors

ANGIREKULA SINDU SRI, GOLLA ARUNA, GOMARI KUSHI, GORRE PRAVALLIKA, MRS. D. GEETHA Author

DOI:

https://doi.org/10.62643/

Keywords:

AI-generated images, deep learning, convolutional neural networks (CNN), gated expert networks, image forensics, fake image detection, generative adversarial networks (GANs), diffusion models, feature extraction, classification, digital security, computer vision.

Abstract

The rapid growth of advanced generative models, such as Generative Adversarial Networks (GANs) and diffusion-based systems, has enabled the creation of highly realistic synthetic images that are increasingly difficult to distinguish from real ones. This raises significant concerns in areas such as misinformation, digital forensics, and content authenticity. To address this challenge, this work proposes a detection framework based on Gated Expert Convolutional Neural Networks (GECNN). The model integrates multiple expert CNNs, each specialized in capturing distinct visual artifacts, along with a gating network that dynamically assigns weights to their outputs based on the input image. This adaptive feature selection mechanism enhances the model’s ability to generalize across various types of AI-generated images. Experimental evaluation demonstrates that the proposed approach achieves higher accuracy and robustness compared to conventional single CNN models, making it effective for real-world applications in image authenticity verification and security systems.