Source Camera Identification in Real Practice: A Preliminary Experimentation

In this paper, an experimental evaluation of one of the most effective source camera identification techniques proposed so far, by Lukáš et al., is presented. This method uses the characteristic noise left by the sensor on a digital picture as a fingerprint in order to identify the source camera used to take the picture. The aim of the experiments is to assess the effectiveness of this technique when used with pictures that were previously modified using several common image-processing functions coming with photo-editing tools. The results seem to confirm that, in most cases, the method by Lukáš et al. is resilient to the modifications introduced by the considered image-processing functions. However, it was possible to identify several cases where the quality of the identification process deteriorated because of the noise introduced by the image-processing.