Semantic Segmentation Based Copy Paste Forgery Detection Anlamsal B l tleme Tabanli Kopyala Yapiştir Sahteciligi Tespiti


Aslantas A. Y., Köylü F.

2023 Innovations in Intelligent Systems and Applications Conference, ASYU 2023, Sivas, Turkey, 11 - 13 October 2023 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/asyu58738.2023.10296832
  • City: Sivas
  • Country: Turkey
  • Keywords: Copy-Move Forgery Detection, Deep Learning, Semantic Segmentation, İmage Forensic, İnformation Security
  • Erciyes University Affiliated: Yes

Abstract

Semantic segmentation refers to the process of finding meaningful sections in a picture. Semantic segmentation is employed in a variety of domains, including the military, robotics, and biomedicine. A new copy-paste forgery approach based on semantic segmentation is provided in this research. Forgery zones were found with improved accuracy and in less time using discrete cosine transforms and semantic segmentation. Because the suggested technique uses semantic segmentation, each region is compared inside its own class, resulting in both time savings and increased performance.