Abstract
Visual messages play a key role in media communication. Modifying and falsifying these messages allows easy manipulation of people's opinions and decisions. The presented work aimed to develop and analyse the convolutional network committee approach to detect fake visual information. The development of artificial intelligence has led to an increase in the sophistication of the reliability of fake images. This phenomenon generates the need for more accurate detection of modified images. For this purpose, the convolutional network committee was created – this includes the exploration of different convolutional network architectures and committee techniques aimed at improving the accuracy and reliability of detecting fake information.
Recommended Citation
Poniszewska-Marańda, Aneta; Pacyniak, Weronika; and Hasimi, Lumbardha, "Enhanced detection of visual fake information via convolutional neural network committee: an analytical study" (2025). Proceedings of the 2025 Pre-ICIS SIGDSA Symposium. 80.
https://aisel.aisnet.org/sigdsa2025/80