Detection And Localization Of Splicing Attacks On Videos Using Block Correlation

Abstract

Temporal attacks (such as copy-move and splicing attacks) are some of the most common tampering techniques in digital video due to their simplicity. While copy-move attacks have been widely studied and different approaches for this task have been proposed in recent years, splicing attacks have been less studied and the works on splicing detection are almost only based on exploiting the sensor pattern noise. This work proposes a straightforward yet effective approach to detect splicing attacks on videos using peak detection over temporal correlation of blocks. This approach allows to detect both the spatial and temporal location of the splicing attacks even when the two videos used in the attack were recorded with the same camera, a case that cannot be addressed with those approaches based on noise patterns.