Source Camera Identification Using Non-decimated Wavelet Transform

Abstract

Source Camera identification of digital images can be performed by matching the sensor pattern noise (SPN) of the images with that of the camera reference signature. This paper presents a non-decimated wavelet based source camera identification method for digital images. The proposed algorithm applies a non-decimated wavelet transform on the input image and split the image into its wavelet sub-bands. The coefficients within the resulting wavelet high frequency sub-bands are filtered to extract the SPN of the image. Cross correlation of the image SPN and the camera reference SPN signature is then used to identify the most likely source device of the image. Experimental results were generated using images of ten cameras to identify the source camera of the images. Results show that the proposed technique generates superior results to that of the state of the art wavelet based source camera identification.