EXISTING RESEARCH, METHODS FOR CREATING FAKE IMAGES AND WAYS TO DETECT THEM

Eremuk Vladimir Vadimovich, Romashov Viktor Andreevich

With the development of advanced technologies in the field of programming, the possibility of creating fake images has arisen, which raises growing concerns about the authenticity of visual content. On the other hand, the detection of fake images has become essential to prevent the spread of misinformation and malicious intent. This thesis paper explores the algorithms used to create fake images and the algorithms used to detect them. Researchers have developed a variety of technical tools, including Photoshop software and deep learning algorithms, to generate fake images with a high degree of realism. At the same time, many detection methods have emerged, including metadata analysis, the use of computer vision algorithms, and the application of blockchain technology to ensure the authenticity of the image. According to recent research, deep learning is a promising approach for developing more accurate anomaly detection algorithms in images. As a consequence, it is critical to develop and continually improve detection methods to ensure the accuracy and authenticity of visual content.

Keywords:fake images, detection methods, computer vision algorithms, blockchain technology, accuracy, deep learning