To obtain access to full text of journal and articles you must register!
- Article name
- About the possibility of AI-generated and real facial images detection by natural and artificial intelligence
- Authors
- Zhumazhanova S. S., , samal_shumashanova@mail.ru, Omsk State Technical University, Omsk, Russia
Grutsyn I. S., , ilya.grutsyn1@gmail.com, Omsk State Technical University, Omsk, Russia
Shemet A. O., , tsemet@inbox.ru, Omsk State Technical University, Omsk, Russia
Agapitov A. V., , avagapitov@mail.ru, Omsk State Technical University, Omsk, Russia
- Keywords
- artificial intelligence / generative artificial intelligence / generative adversarial networks / diffusion models / explainable artificial intelligence / information privacy / synthesized content / deepfake
- Year
- 2026 Issue 2 Pages 49 - 59
- Code EDN
- DNYBGF
- Code DOI
- 10.52190/2073-2600_2026_2_49
- Abstract
- With the rapid development of generative neural network models, the task of distinguishing between real and synthetic facial images is becoming critical for information security. This article compares the effectiveness of generated content detection using natural (human) and artificial intelligence (methods based on spectral features of facial images). The accuracy of both approaches is assessed using field and computational experiments. The results demonstrates the superiority of automated systems in detecting generated images, but emphasize the role of human expertise in contextual analysis. Prospects for creating hybrid transparent (explainable) systems to minimize the risks of fraud and the spread of disinformation are discussed.
- Text
- BUY for read the full text of article
- Buy
- 500.00 rub
