Student:	Işılay BOZKURT
Supervisor:	Yrd. Doç. Dr. Güzin ULUTAŞ
Department:	Bilgisayar Mühendisliği
Institution:	Graduate School of Natural and Applied Sciences
University:	Karadeniz Technical University Turkey

Title of the Thesis:	FRAME DUPLİCATİON DETECTİON FORGERY İN VİDEOS
Level:	M.Sc.
Acceptance Date:	2/1/2017
Number of Pages:	68
Registration Number:	i3149

Summary:

Malicious users can edit videos for hiding an object or incident. These people can make

forgery as frame deletion from videos, insertion frames into videos and duplication some of

frames in the videos. The changes made on the video leave statistical traces in the video.

Forgery detection is done from these traces.

In this thesis scope, frame duplication forgery detection in videos is studied. The current

methods in literature is investigated, new methods that robust against compression attacks is

suggested, and experimental results are reported. The proposed methods perform effective

forgery detection and provide faster execution time compared to other methods in the literature.

Keywords: Video Forgery, Frame Duplication, Color Layout Descriptor, Peak Signal-to-Noise

Ratio, Discrete Cosines Transform, Hough Transform