Author : Anupam Shukla 1
Date of Publication :22nd June 2017
Abstract: An enormous amount of images or videos are collected from laptops, mobiles, storage devices during the investigation by Police or intelligence agencies or digital forensic team. These collected images/videos to be analyzed to ascertain the source device that was used to capture these during the investigation. An every camera has its fingerprint in the form of Photo Response Non- Uniformity (PRNU) noise. Because it has universality and generality nature, it is unique and hence plays a very vital role in the source camera identification. PRNU is a sensor pattern noise which contains noise components, and other information hence many techniques have been proposed for the extraction of the PRNU. In this paper, a Discrete Cosine Transform (DCT) method is used for extracting the noise, and Weighted Averaging technique for PRNU estimation. Finally, the distance function is used for comparing g the difference between the Query image PRNU and the stored image PRNU. We conducted the experiments its results are verified against the different cameras, and it is giving 93% accuracy
Reference :