Video Forgery Detection Based on Variance in Luminance and Signal to Noise Ratio using LESH Features and Bispectral Analysis?

One among the most popular topics today is authentication of the digital content one has. Due to easy availability of advanced tools to modify content it has become difficult to judge whether a given content produced as piece of evidence is valid or forged one. The main aim of this paper to introduce a method based on statistical properties useful for detecting the forgeries in digital content specially videos. In this the key properties such as luminance and signal to noise ratio has been explored and use of Local Energy based shape histogram and Bispectral analysis. This paper mainly proposes a method which proves useful in detecting the different forms of tampering done with videos such copy-paste, inpainting etc. Experimental results shows that proposed method works in an efficient manner for videos tampered using different techniques.