Detection of frame duplication type of forgery in digital video using sub-block based features

Abstract

With the easy availability and operability of video editing tools, any video could be edited in short span of time. Sometimes, these modifications change the actual meaning of targeted video. Hence, before making any judgment and opinion about such multimedia contents, it is necessary to verify their genu‐ ineness. A video can be tampered by various different attempts. Each different attempt derives a new type of forgery in videos. Among various types of attack on video, frame duplication is a common type of attack. Frames are duplicated and pasted into same video in order to either hide or add false information. We propose Sub Blocked based features to detect frame duplication. The experi‐ mental results show higher accuracy that not only detects but also localize dupli‐ cated frames as well.