Forensic Camera Classification: Verification of Sensor Pattern Noise Approach

Digital images can be obtained through a variety of sources including digital cameras and scanners. In many cases the ability to determine the source of a digital image is important. This paper verifies the method presented in (Lukas et al. 2005) for authenticating images that have been acquired using digital still cameras. The method is based on the usage of imaging sensor pattern noise. A reference pattern is estimated for each camera and is treated as a unique fingerprint of that camera. To identify the source camera of an unknown image, the noise extracted from the image is correlated with all the reference camera patterns obtained from training images. This correlation based approach is shown to have close to 100% classification accuracy.